Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/icce53296.2022.9730332

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：ASTES Journal  

DOI： 10.25046/aj060519

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA  

The biggest obstacle to develop end-to-end Japanese text-to-speech (TTS) systems is to estimate phonetic and prosodic information (PPI) from Japanese texts. The following are the reasons: (1) the Kanji characters of the Japanese writing system have multiple corresponding pronunciations, (2) there is no separation mark between words, and (3) an accent nucleus must be assigned at appropriate positions. In this paper, we propose to solve the problems by neural machine translation (NMT) on the basis of encoder-decoder models, and compare NMT models of recurrent neural networks and the Transformer architecture. The proposed model handles texts on token (character) basis, although conventional systems handle them on word basis. To ensure the potential of the proposed approach, NMT models are trained using pairs of sentences and their PPIs that are generated by a conventional Japanese TTS system from 5 million sentences. Evaluation experiments were performed using PPIs that are manually annotated for 5,142 sentences. The experimental results showed that the Transformer architecture has the best performance, with 98.0% accuracy for phonetic information estimation and 95.0% accuracy for PPI estimation. Judging from the results, NMT models are promising toward end-to-end Japanese TTS.

DOI： 10.21437/interspeech.2021-914

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Elsevier BV  

This paper proposes architectures that facilitate the extrapolation of emotional expressions in deep neural network (DNN)-based text-to-speech (TTS). In this study, the meaning of “extrapolate emotional expressions” is to borrow emotional expressions from others, and the collection of emotional speech uttered by target speakers is unnecessary. Although a DNN has potential power to construct DNN-based TTS with emotional expressions and some DNN-based TTS systems have demonstrated satisfactory performances in the expression of the diversity of human speech, it is necessary and troublesome to collect emotional speech uttered by target speakers. To solve this issue, we propose architectures to separately train the speaker feature and the emotional feature and to synthesize speech with any combined quality of speakers and emotions. The architectures are parallel model (PM), serial model (SM), auxiliary input model (AIM), and hybrid models (PM&AIM and SM&AIM). These models are trained through emotional speech uttered by few speakers and neutral speech uttered by many speakers. Objective evaluations demonstrate that the performances in the open-emotion test provide insufficient information. They make a comparison with those in the closed-emotion test, but each speaker has their own manner of expressing emotion. However, subjective evaluation results indicate that the proposed models could convey emotional information to some extent. Notably, the PM can correctly convey sad and joyful emotions at a rate of 60%.

DOI： 10.1016/j.specom.2020.11.004

Web of Science

researchmap

Other Link： https://arxiv.org/abs/2102.10345

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE/APSIPA  

End-to-end text-to-speech (TTS) models have achieved remarkable results in recent times. However, the model requires a large amount of text and audio data for training. A speaker adaptation method based on fine-tuning has been proposed for constructing a TTS model using small scale data. Although these methods can replicate the target speaker s voice quality, synthesized speech includes the deletion and/or repetition of speech. The goal of speaker adaptation is to change the voice quality to match the target speaker ' s on the premise that adjusting the necessary modules will reduce the amount of data to be fine-tuned. In this paper, we clarify the role of each module in the Transformer-TTS process by not updating it. Specifically, we froze character embedding, encoder, layer predicting stop token, and loss function for estimating sentence ending. The experimental results showed the following: (1) fine-tuning the character embedding did not result in an improvement in the deletion and/or repetition of speech, (2) speech deletion increases if the encoder is not fine-tuned, (3) speech deletion was suppressed when the layer predicting stop token is not fine-tuned, and (4) there are frequent speech repetitions at sentence end when the loss function estimating sentence ending is omitted.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/tencon50793.2020.9293766

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA  

DOI： 10.21437/interspeech.2020-2064

researchmap

Other Link： https://www.webofscience.com/wos/woscc/full-record/WOS:000833594103112

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Recently, end-to-end text-to-speech (TTS) models have achieved a remarkable performance, however, requiring a large amount of paired text and speech data for training. On the other hand, we can easily collect unpaired dozen minutes of speech recordings for a target speaker without corresponding text data. To make use of such accessible data, the proposed method leverages the recent great success of state-of-the-art end-to-end automatic speech recognition (ASR) systems and obtains corresponding transcriptions from pretrained ASR models. Although these models could only provide text output instead of intermediate linguistic features like phonemes, end-to-end TTS can be well trained with such raw text data directly. Thus, the proposed method can greatly simplify a speaker adaptation pipeline by consistently employing end-to-end ASR/TTS ecosystems. The experimental results show that our proposed method achieved comparable performance to a paired data adaptation method in terms of subjective speaker similarity and objective cepstral distance measures.

DOI： 10.1109/icassp40776.2020.9053371

Web of Science

researchmap

Authorship：Corresponding author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we propose using phonemic information in addition to acoustic features to improve the intelligibility of speech uttered by patients with articulation disorders caused by a wide glossectomy. Our previous studies showed that voice conversion algorithm improves the quality of glossectomy patients' speech. However, losses in acoustic features of glossectomy patients' speech are so large that the quality of the reconstructed speech is low. To solve this problem, we explored potentials of several additional information to improve speech intelligibility. One of the candidates is phonemic information, more specifically Phoneme Labels as Auxiliary input (PLA). To combine both acoustic features and PLA, we employed a DNN-based algorithm. PLA is represented by a kind of one-of-k vector, i.e., PLA has a weight value (<1.0) that gradually changes in time axis, whereas one-of-k has a binary value (0 or 1). The results showed that the proposed algorithm reduced the mel-frequency cepstral distortion for all phonemes, and almost always improved intelligibility. Notably, the intelligibility was largely improved in phonemes /s/ and /z/, mainly because the tongue is used to sustain constriction to produces these phonemes. This indicates that PLA works well to compensate the lack of a tongue.

DOI： 10.1109/APSIPAASC47483.2019.9023168

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we propose a new algorithm to generate Speech-like Emotional Sound (SES). Emotional information plays an important role in human communication, and speech is one of the most useful media to express emotions. Although, in general, speech conveys emotional information as well as linguistic information, we have undertaken the challenge to generate sounds that convey emotional information without linguistic information, which results in making conversations in human-machine interactions more natural in some situations by providing non-verbal emotional vocalizations. We call the generated sounds "speech-like", because the sounds do not contain any linguistic information. For the purpose, we propose to employ WaveNet as a sound generator conditioned by only emotional IDs. The idea is quite different from WaveNet Vocoder that synthesizes speech using spectrum information as auxiliary features. The biggest advantage of the idea is to reduce the amount of emotional speech data for the training. The proposed algorithm consists of two steps. In the first step, WaveNet is trained to obtain phonetic features using a large speech database, and in the second step, WaveNet is re-trained using a small amount of emotional speech. Subjective listening evaluations showed that the SES could convey emotional information and was judged to sound like a human voice.

DOI： 10.1109/APSIPAASC47483.2019.9023346

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER INTERNATIONAL PUBLISHING AG  

This study focuses on gesture recognition in mobile interaction settings, i.e. when the interacting partners are walking. This kind of interaction requires a particular coordination, e.g. by staying in the field of view of the partner, avoiding obstacles without disrupting group composition and sustaining joint attention during motion. In literature, various studies have proven that gestures are in close relation in achieving such goals.Thus, a mobile robot moving in a group with human pedestrians, has to identify such gestures to sustain group coordination. However, decoupling of the inherent -walking- oscillations and gestures, is a big challenge for the robot. To that end, we employ video data recorded in uncontrolled settings and detect arm gestures performed by human-human pedestrian pairs by adopting a signal processing approach. Namely, we exploit the fact that there is an inherent oscillatory motion at the upper limbs arising from the gait, independent of the view angle or distance of the user to the camera. We identify arm gestures as disturbances on these oscillations. In doing that, we use a simple pitch detection method from speech processing and assume data involving a low frequency periodicity to be free of gestures. In testing, we employ a video data set recorded in uncontrolled settings and show that we achieve a detection rate of 0.80.

DOI： 10.1007/978-3-030-26118-4_8

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA  

In this paper, we propose an algorithm to improve the naturalness of the reconstructed glossectomy patient's speech that is generated by voice conversion to enhance the intelligibility of speech uttered by patients with a wide glossectomy. While existing VC algorithms make it possible to improve intelligibility and naturalness, the result is still not satisfying. To solve the continuing problems, we propose to directly modify the speech waveforms using a spectrum differential. The motivation is that glossectomy patients mainly have problems in their vocal tract, not in their vocal cords. The proposed algorithm requires no source parameter extractions for speech synthesis, so there are no errors in source parameter extractions and we are able to make the best use of the original source characteristics. In terms of spectrum conversion, we evaluate with both GMM and DNN. Subjective evaluations show that our algorithm can synthesize more natural speech than the vocoder-based method. Judging from observations of the spectrogram, power in high-frequency bands of fricatives and stops is reconstructed to be similar to that of natural speech.

DOI： 10.21437/Interspeech.2018-1239

Web of Science

researchmap

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Sounds are one of the most valuable information sources for human beings from the viewpoint of understanding the environment around them. We have been now investigating the method of detecting and visualizing crowded situations in the city in a sound-sensing manner. For this purpose, we have developed a sound collection system oriented to a crowdsourcing approach and carried out the sound-collection in two Japanese cities, Okayama and Kurashiki. In this paper, we present an overview of sound collections. Then, to show an effectiveness of analyzation by sensed sounds, we profile characteristics of the cities through the visualization results of the sound.

DOI： 10.1109/APSIPA.2017.8282238

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we investigate deep neural network (DNN) architectures to transplant emotional expressions to improve the expressiveness of DNN-based text-to-speech (TTS) synthesis. DNN is expected to have potential power in mapping between linguistic information and acoustic features. From multispeaker and/or multi-language perspectives, several types of DNN architecture have been proposed and have shown good performances. We tried to expand the idea to transplant emotion, constructing shared emotion-dependent mappings. The following three types of DNN architecture are examined; (1) the parallel model (PM) with an output layer consisting of both speaker-dependent layers and emotion-dependent layers, (2) the serial model (SM) with an output layer consisting of emotion-dependent layers preceded by speaker-dependent hidden layers, (3) the auxiliary input model (AIM) with an input layer consisting of emotion and speaker IDs as well as linguistics feature vectors. The DNNs were trained using neutral speech uttered by 24 speakers, and sad speech and joyful speech uttered by 3 speakers from those 24 speakers. In terms of unseen emotional synthesis, subjective evaluation tests showed that the PM performs much better than the SM and slightly better than the AIM. In addition, this test showed that the SM is the best of the three models when training data includes emotional speech uttered by the target speaker.

DOI： 10.1109/APSIPA.2017.8282231

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ACM  

DOI： 10.1145/3148150.3148151

researchmap

Other Link： http://doi.acm.org/10.1145/3148150.3148151

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ACM  

In this paper, we investigate a method of creating noise maps that take account of human senses. Physical measurements are not enough to design our living environment and we need to know subjective assessments. To predict subjective assessments from loudness values, we propose to use metadata related to where, who and what is recording. The proposed method is implemented using deep neural networks because these can naturally treat a variety of information types. First, we evaluated its performance in predicting five-point subjective loudness levels based on a combination of several features: location specific, participant-specific, and sound-specific features. The proposed method achieved a 16.3 point increase compared with the baseline method. Next, we evaluated its performance based on noise map visualization results. The proposed noise maps were generated from the predicted subjective loudness level. Considering the differences between the two visualizations, the proposed method made fewer errors than the baseline method.

DOI： 10.1145/3123024.3123091

Web of Science

researchmap

Other Link： http://doi.acm.org/10.1145/3123024.3123091

Publishing type：Research paper (international conference proceedings)   Publisher：ISCA  

DOI： 10.21437/interspeech.2017-841

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA-INT SPEECH COMMUNICATION ASSOC  

In this paper, using GMM-based voice conversion algorithm, we propose to generate speaker-dependent mapping functions to improve the intelligibility of speech uttered by patients with a wide glossectomy. The speaker-dependent approach enables to generate the mapping functions that reconstruct missing spectrum features of speech uttered by a patient without having influences of a speaker's factor. The proposed idea is simple, i.e., to collect speech uttered by a patient before and after the glossectomy, but in practice it is hard to ask patients to utter speech just for developing algorithms. To confirm the performance of the proposed approach, in this paper, in order to simulate glossectomy patients, we fabricated an intraoral appliance which covers lower dental arch and tongue surface to restrain tongue movements. In terms of the Mel-frequency cepstrum (MFC) distance, by applying the voice conversion, the distances were reduced by 25% and 42% for speaker dependent case and speaker-independent case, respectively. In terms of phoneme intelligibility, dictation tests revealed that speech reconstructed by speaker-dependent approach almost always showed better performance than the original speech uttered by simulated patients, while speaker-independent approach did not.

DOI： 10.21437/Interspeech.2017-841

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/apsipa.2016.7820909

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computing Machinery, Inc  

We propose a lifelog browsing system through which users can share memories of their experiences with other users. Most importantly, by using global positioning system data and time stamps, the system simultaneously displays a variety of log information in a time-synchronous manner. This function empowers users with not only an easy interpretation of other users' experiences but also nonverbal notifications. Shared information on this system includes photographs taken by users, Google street views, shops and restaurants on the map, daily weather, and other items relevant to users' interests. In evaluation experiments, users preferred the proposed system to conventional photograph albums and maps for explaining and sharing their experiences. Moreover, through displayed information, the listeners found out their interest items that had not been mentioned by the speakers.

DOI： 10.1145/2968219.2971401

Scopus

researchmap

Other Link： http://doi.acm.org/10.1145/2968219.2971401

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computing Machinery, Inc  

In this study, in order to develop a monitoring system that takes into account privacy issues, we investigated user preferences in terms of the monitoring and privacy protec-tion levels. The people on the monitoring side wanted the monitoring system to allow them to monitor in detail. Con-versely, it was observed for the people being monitored that the more detailed the monitoring, the greater the feelings of being surveilled intrusively. Evaluation experiments were performed using the location data of three people in differ-ent living areas. The results of the experiments show that it is possible to control the levels of monitoring and privacy protection without being affected by the shape of a living area by adjusting the quantization level of location informa-tion. Furthermore, it became clear that the granularity of location information satisfying the people on the monitored side and the monitoring side is different.

DOI： 10.1145/2968219.2971412

Scopus

researchmap

Other Link： http://doi.acm.org/10.1145/2968219.2971412

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents a sound collection system that uses crowdsourcing to gather information for visualizing area characteristics. First, we developed a sound collection system to simultaneously collect physical sounds, their statistics, and subjective evaluations. We then conducted a sound collection experiment using the developed system on 14 participants. We collected 693,582 samples of equivalent Aweighted loudness levels and their locations, and 5,935 samples of sounds and their locations. The data also include subjective evaluations by the participants. In addition, we analyzed the changes in sound properties of some areas before and after the opening of a large-scale shopping mall in a city. Next, we implemented visualizations on the server system to attract users' interests. Finally, we published the system, which can receive sounds from any Android smartphone user. The sound data were continuously collected and achieved a specified result.

DOI： 10.1109/ICMEW.2016.7574694

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

We propose a spoken dialog strategy for car navigation systems to facilitate safe driving. To drive safely, drivers need to concentrate on their driving; however, their concentration may be disrupted due to disagreement with their spoken dialog system. Therefore, we need to solve the problems of user misunderstandings as well as misunderstanding of spoken dialog systems. For this purpose, we introduced a driver workload level in spoken dialog management in order to prevent user misunderstandings. A key strategy of the dialog management is to make speech redundant if the driver's workload is too high in assuming that the user probably misunderstand the system utterance under such a condition. An experiment was conducted to compare performances of the proposed method and a conventional method using a user simulator. The simulator is developed under the assumption of two types of drivers: an experienced driver model and a novice driver model. Experimental results showed that the proposed strategies achieved better performance than the conventional one for task completion time, task completion rate, and user's positive speech rate. In particular, these performance differences are greater for novice users than for experienced users.

DOI： 10.1109/APSIPA.2015.7415511

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents an algorithm to mine basic patterns of human activities on a daily basis using non-negative matrix factorization (NMF). The greatest benefit of the algorithm is that it can elicit patterns from which meanings can be easily interpreted. To confirm its performance, the proposed algorithm was applied to PC logging data collected from three occupations in offices. Daily patterns of software usage were extracted for each occupation. Results show that each occupation uses specific software in its own time period, and uses several types of software in parallel in its own combinations. Experiment results also show that patterns of 144 dimension vectors were compressible to those of 11 dimension vectors without degradation in occupation classification performance. Therefore, the proposed algorithm compressed basic software usage patterns to about one-tenth of their original dimensions while preserving the original information. Moreover, the extracted basic patterns showed reasonable interpretation of daily working patterns in offices.

DOI： 10.1109/ICCE.2015.7066309

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA  

As described in this paper, we propose a sub-band speech synthesis approach to develop a high quality Text-to-Speech (TTS) system: a sample-based spectrum is used in the high-frequency band and spectrum generated by HMM-based TTS is used in the low-frequency band. Herein, sample-based spectrum means spectrum selected from a phoneme database such that it is the most similar to spectrum generated by HMM-based speech synthesis. A key idea is to compensate over-smoothing caused by statistical procedures by introducing a sample-based spectrum, especially in the high-frequency band. Listening test results show that the proposed method has better performance than HMM-based speech synthesis in terms of clarity. It is at the same level as HMM-based speech synthesis in terms of smoothness. In addition, preference test results among the proposed method, HMM-based speech synthesis, and waveform speech synthesis using 80 min speech data reveal that the proposed method is the most liked.

Web of Science

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/interspeech/interspeech2015.html#conf/interspeech/InaiHAIMM15

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents a sound collection system to visualize environmental sounds that are collected using a crowd sourcing approach. An analysis of physical features is generally used to analyze sound properties; however, human beings not only analyze but also emotionally connect to sounds. If we want to visualize the sounds according to the characteristics of the listener, we need to collect not only the raw sound, but also the subjective feelings associated with them. For this purpose, we developed a sound collection system using a crowdsourcing approach to collect physical sounds, their statistics, and subjective evaluations simultaneously. We then conducted a sound collection experiment using the developed system on ten participants. We collected 6,257 samples of equivalent loudness levels and their locations, and 516 samples of sounds and their locations. Subjective evaluations by the participants are also included in the data. Next, we tried to visualize the sound on a map. The loudness levels are visualized as a color map and the sounds are visualized as icons which indicate the sound type. Finally, we conducted a discrimination experiment on the sound to implement a function of automatic conversion from sounds to appropriate icons. The classifier is trained on the basis of the GMM-UBM (Gaussian Mixture Model and Universal Background Model) method. Experimental results show that the F-measure is 0.52 and the AUC is 0.79.

DOI： 10.1109/PERCOMW.2015.7134069

Web of Science

researchmap

Other Link： https://ousar.lib.okayama-u.ac.jp/ja/53271

Language：English   Publishing type：Part of collection (book)   Publisher：SPRINGER-VERLAG BERLIN  

We propose an algorithm to estimate a person's living area using his/her collected Global Positioning System (GPS) data. The most important feature of the algorithm is the connectivity of places with a home, i.e., a living area must consist of a home, important places, and routes that connect them. This definition is logical because people usually go to a place from home, and there can be several routes to that place. Experimental results show that the proposed algorithm can estimate living area with a precision of 0.82 and recall of 0.86 compared with the grand truth established by users. It is also confirmed that the connectivity of places with a home is necessary to estimate a reasonable living area.

DOI： 10.1007/978-3-319-21383-5_95

Web of Science

researchmap

Language：English   Publishing type：Part of collection (book)   Publisher：SPRINGER-VERLAG BERLIN  

We propose a method to extract particular sound segments from the sound recorded during the course of a day in order to provide sound segments that can be used to facilitate memory. To extract important parts of the sound data, the proposed method utilizes human behavior based on a multisensing approach. To evaluate the performance of the proposed method, we conducted experiments using sound, acceleration, and global positioning system data collected by five participants for approximately two weeks. The experimental results are summarized as follows: (1) various sounds can be extracted by dividing a day into scenes using the acceleration data; (2) sound recorded in unusual places is preferable to sound recorded in usual places; and (3) speech is preferable to nonspeech sound.

DOI： 10.1007/978-3-319-21380-4_105

Web of Science

researchmap

Language：English   Publisher：W B SAUNDERS CO-ELSEVIER INC  

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Information and Media Technologies Editorial Board  

We regarded a dialog strategy for information retrieval as a graph search problem and proposed several novel dialog strategies that can recover from misrecognition through a spoken dialog that traverses the graph. To recover from misrecognition without seeking confirmation, our system kept multiple understanding hypotheses at each turn and searched for a globally optimal hypothesis in the graph whose nodes express understanding states across user utterances in a whole dialog. In the search, we used a new criterion based on efficiency in information retrieval and consistency with understanding hypotheses, which is also used to select an appropriate system response. We showed that our system can make more efficient and natural dialogs than previous ones.

DOI： 10.11185/imt.9.111

CiNii Article

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

We are striving to develop a new communication technology based on individual experiences that can be extracted from life logs. We have proposed the "Emotion Communication Model" and confirmed that significant correlation exists between experience and emotion. As the second step, particularly addressing impressive places and events, this paper describes an investigation of the extent to which we can share emotional information with others through individuals' experiences. Subjective experiments were conducted using life log data collected during 7-47 months. Experiment results show that (1) impressive places are determined by the distance from home, visit frequency, and direction from home and that (2) positive emotional information is highly consistent among people (71.4%), but it is not true for negative emotional information. Therefore, experiences are useful as metaphors to express positive emotional information.

DOI： 10.1109/PerComW.2014.6815198

researchmap

Authorship：Lead author   Language：English   Publishing type：Part of collection (book)   Publisher：Springer  

DOI： 10.1007/978-1-4614-8280-2_8

researchmap

Language：English   Publishing type：Part of collection (book)   Publisher：Springer  

DOI： 10.1007/978-1-4614-8280-2_35

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper proposes a sub-band speech synthesis approach to develop high-quality Text-to-Speech (TTS). For the low-frequency band and high-frequency band, Hidden Markov Model (HMM)-based speech synthesis and waveform -based speech synthesis are used, respectively. Both speech synthesis methods are widely known to show good performance and to have benefits and shortcomings from different points of view. One motivation is to apply the right speech synthesis method in the right frequency band. Experiment results show that in terms of the smoothness the proposed approach shows better performance than waveform -based speech synthesis, and in terms of the clarity it shows better than HMM-based speech synthesis. Consequently, the proposed approach combines the inherent benefits from both waveform-based speech synthesis and HMM-based speech synthesis.

DOI： 10.1109/APSIPA.2014.7041575

Web of Science

researchmap

Language：Japanese  

On a real environment speech-oriented information guidance system, a valid and invalid input discrimination is important as invalid inputs such as noise, laugh, cough and utterances between users lead to unpredictable system responses. Generally, acoustic features such as MFCC (Mel-Frequency Cepstral Coefficient) are used for discrimination. Comparing acoustic likelihoods of GMMs (Gaussian Mixture Models) from speech data and noise data is one of the typical methods. In addition to that, using linguistic features, such as speech recognition result, is considered to improve discrimination accuracy as it reflects the task-domain of invalid inputs and meaningless recognition results from noise inputs. In this paper, we introduce Bag-of-Words (BOW) as a feature to discriminate between valid and invalid inputs. Support Vector Machine (SVM) and Maximum Entropy method (ME) are also employed to realize robust classification. We experimented the methods using real environment data obtained from the guidance system "Takemaru-kun." By applying BOW on SVM, the F-measure is improved to 85.09%, from 82.19% when using GMMs. In addition, experiments using features combining BOW with acoustic likelihoods from GMMs, Duration and SNR were conducted, improving the F-measure to 86.58%.

CiNii Article

CiNii Books

researchmap

Authorship：Lead author   Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Other Link： http://search.ieice.org/bin/summary.php?id=j96-d_1_81&category=D&year=2013&lang=J&abst=

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

A novel speech feature generation-based acoustic model training method for robust speaker-independent speech recognition is proposed. For decades, speaker adaptation methods have been widely used. All of these adaptation methods need adaptation data. However, our proposed method aims to create speaker-independent acoustic models that cover not only known but also unknown speakers. We achieve this by adopting inverse maximum likelihood linear regression (MLLR) transformation-based feature generation, and then we train our models using these features. First we obtain MLLR transformation matrices from a limited number of existing speakers. Then we extract the bases of the MLLR transformation matrices using PCA. The distribution of the weight parameters to express the transformation matrices for the existing speakers are estimated. Next, we construct pseudo-speaker transformations by sampling the weight parameters from the distribution, and apply the transformation to the normalized features of the existing speaker to generate the features of the pseudo-speakers. Finally, using these features, we train the acoustic models. Evaluation results show that the acoustic models trained using our proposed method are robust for unknown speakers.

DOI： 10.1587/transinf.E95.D.2479

Web of Science

researchmap

Other Link： http://search.ieice.org/bin/summary.php?id=e95-d_10_2479

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：EUROPEAN LANGUAGE RESOURCES ASSOC-ELRA  

In this paper, we analyze the causes of task completion errors in spoken dialog systems, using a decision tree with N-gram features of the dialog to detect task-incomplete dialogs. The dialog for a music retrieval task is described by a sequence of tags related to user and system utterances and behaviors. The dialogs are manually classified into two classes: completed and uncompleted music retrieval tasks. Differences in tag classification performance between the two classes are discussed. We then construct decision trees which can detect if a dialog finished with the task completed or not, using information gain criterion. Decision trees using N-grams of manual tags and automatic tags achieved 74.2% and 80.4% classification accuracy, respectively, while the tree using interaction parameters achieved an accuracy rate of 65.7%. We also discuss more details of the causality of task incompletion for spoken dialog systems using such trees.

Web of Science

researchmap

Other Link： http://www.lrec-conf.org/proceedings/lrec2012/summaries/1059.html

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we propose a novel acoustic model training method which is suitable for speaker adaptation in speech recognition. Our method is based on feature generation from a small amount of speakers' data. For decades, speaker adaptation methods have been widely used. Such adaptation methods need some amount of adaptation data and if the data is not sufficient, speech recognition performance degrade significantly. If the seed models to be adapted to a specific speaker can widely cover more speakers, speaker adaptation can perform robustly. To make such robust seed models, we adopt inverse maximum likelihood linear regression (MLLR) transformation-based feature generation, and then train our seed models using these features. First we obtain MLLR transformation matrices from a limited number of existing speakers. Then we extract the bases of the MLLR transformation matrices using PCA. The distribution of the weight parameters to express the MLLR transformation matrices for the existing speakers is estimated. Next we generate pseudo-speaker MLLR transformations by sampling the weight parameters from the distribution, and apply the inverse of the transformation to the normalized existing speaker features to generate the pseudo-speakers' features. Finally, using these features, we train the acoustic seed models. Using this seed models, we obtained better speaker adaptation results than using simply environmentally adapted models. © 2011 IEEE.

DOI： 10.1109/ASRU.2011.6163925

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：APSIPA  

In this paper a novel speech feature generationbased acoustic model training method is proposed. For decades, speaker adaptation methods have been widely used. All existing adaptation methods need adaptation data. However, our proposed method creates speaker-independent acoustic models that cover not only known but also unknown speakers. We do this by adopting inverse maximum likelihood linear regression (MLLR) transformation-based feature generation, and then train our models using these features. First we obtain MLLR transformation matrices from a limited number of existing speakers. Then we extract the bases of the MLLR transformation matrices using PCA. The distribution of the weight parameters to express the MLLR transformation matrices for the existing speakers are estimated. Next we generate pseudo-speaker MLLR transformations by sampling the weight parameters from the distribution, and apply the inverse of the transformation to the normalized existing speaker features to generate the pseudospeakers' features. Finally, using these features, we train the acoustic models. Evaluation results show that the acoustic models which are created are robust for unknown speakers.

Scopus

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA-INT SPEECH COMMUNICATION ASSOC  

We propose a method of detecting "task incomplete" dialogs in spoken dialog systems using N-gram-based dialog models. We used a database created during a field test in which inexperienced users used a client-server music retrieval system with a spoken dialog interface on their own PCs. In this study, the dialog for a music retrieval task consisted of a sequence of user and system tags that related their utterances and behaviors. The dialogs were manually classified into two classes: the dialog either completed the music retrieval task or it didn't. We then detected dialogs that did not complete the task, using N-gram probability models or a Support Vector Machine with N-gram feature vectors trained using manually classified dialogs. Off-line and on-line detection experiments were conducted on a large amount of real data, and the results show that our proposed method achieved good classification performance.

Web of Science

researchmap

Other Link： http://www.isca-speech.org/archive/interspeech_2011/i11_1305.html

Language：English   Publishing type：Part of collection (book)   Publisher：IOS PRESS  

We developed an ambient system that plays music suitable for the mood of a human-human conversation using words obtained from a continuous-speech recognition system. Using the correspondence between a document space based on the texts related to the music and an acoustic space that expresses various audio features, the continuous-speech recognition results are mapped to an acoustic space. We performed a subjective evaluation of the system. The subjects rated the recommended music and the result reveals that the 10 most highly recommended selections included suitable music.

DOI： 10.3233/978-1-60750-795-6-352

Web of Science

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ISCA  

In this paper, we propose a method of detecting task-incompleted users for a spoken dialog system using an N-gram-based dialog history model. We collected a large amount of spoken dialog data accompanied by usability evaluation scores by users in real environments. The database was made by a field test in which naive users used a client-server music retrieval system with a spoken dialog interface on their own PCs. An N-gram model was trained from sequences that consist of user dialog acts and/or system dialog acts for two dialog classes, that is, the dialog completed the music retrieval task or the dialog incompleted the task. Then the system detects unknown dialogs that is not completed the task based on the N-gram likelihood. Experiments were conducted on large real data, and the results show that our proposed method achieved good classification performance. When the classifier correctly detected all of the task-incompleted dialogs, our proposed method achieved a false detection rate of 6%.

Web of Science

researchmap

Other Link： http://www.isca-speech.org/archive/interspeech_2010/i10_3034.html

Language：English   Publishing type：Research paper (international conference proceedings)  

We propose a technique for generating a large amount of target speaker-like speech features by converting a large amount of prepared speech features of many speakers into features similar to those of the target speaker using a transformation matrix. To generate a large amount of target speaker-like features, the system only needs a very small amount of the target speaker's utterances. This technique enables the system to adapt the acoustic model efficiently from a small amount of the target speaker's utterances. To evaluate the proposed method, we prepared 100 reference speakers and 12 target (test) speakers. We conducted the experiments in an isolated word recognition task using a speech database collected by real PC-based distributed environments and compared our proposed method with MLLR, MAP and the method theoretically equivalent to the SAT. Experimental results proved that the proposed method needed a significantly smaller amount of the target speaker's utterances than conventional MLLR, MAP and SAT.

Scopus

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：EUROPEAN LANGUAGE RESOURCES ASSOC-ELRA  

In this paper, we propose an estimation method of user satisfaction for a spoken dialog system using an N-gram-based dialog history model. We have collected a large amount of spoken dialog data accompanied by usability evaluation scores by users in real environments. The database is made by a field-test in which naive users used a client-server music retrieval system with a spoken dialog interface on their own PCs. An N-gram model is trained from the sequences that consist of users' dialog acts and/or the system's dialog acts for each one of six user satisfaction levels: from 1 to 5 and phi (task not completed). Then, the satisfaction level is estimated based on the N-gram likelihood. Experiments were conducted on the large real data and the results show that our proposed method achieved good classification performance; the classification accuracy was 94.7% in the experiment on a classification into dialogs with task completion and those without task completion. Even if the classifier detected all of the task incomplete dialog correctly, our proposed method achieved the false detection rate of only 6%.

Web of Science

researchmap

Other Link： http://www.lrec-conf.org/proceedings/lrec2010/summaries/579.html

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：EUROPEAN LANGUAGE RESOURCES ASSOC-ELRA  

In this paper, a large-scale real-world speech database is introduced along with other multimedia driving data. We designed a data collection vehicle equipped with various sensors to synchronously record twelve-channel speech, three-channel video, driving behavior including gas and brake pedal pressures, steering angles, and vehicle velocities, physiological signals including driver heart rate, skin conductance, and emotion-based sweating on the palms and soles, etc. These multimodal data are collected while driving on city streets and expressways under four different driving task conditions including two kinds of monologues, human-human dialog, and human-machine dialog. We investigated the response timing of drivers against navigator utterances and found that most overlapped with the preceding utterance due to the task characteristics and the features of Japanese. When comparing utterance length, speaking rate, and the filler rate of driver utterances in human-human and human-machine dialogs, we found that drivers tended to use longer and faster utterances with more fillers to talk with humans than machines.

Web of Science

researchmap

Other Link： http://www.lrec-conf.org/proceedings/lrec2008/summaries/472.html

Authorship：Lead author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)  

CiNii Article

researchmap

Language：English   Publisher：W B SAUNDERS CO-ELSEVIER INC  

0

Web of Science

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：The Japanese Society for Artificial Intelligence  

DOI： 10.11517/jjsai.28.1_52

CiNii Article

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publishing type：Rapid communication, short report, research note, etc. (scientific journal)   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：日本音響学会関西支部  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：日本顎顔面補綴学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：English   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：電気・情報関連学会中国支部  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：日本音響学会関西支部  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：日本補綴歯科学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Authorship：Corresponding author   Language：English   Publishing type：Research paper, summary (national, other academic conference)   Publisher：日本生物物理学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：一般社団法人電子情報通信学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：IEEE広島支部  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publisher：日本音響学会  

researchmap

Language：Japanese   Publisher：日本音響学会  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese  

CiNii Article

CiNii Books

researchmap

Authorship：Corresponding author   Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese   Publishing type：Research paper, summary (international conference)  

DOI： 10.1121/1.4969721

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

One of the applications using life log is to retrieve memorable scenes. In this paper, for extracting memorable scenes from video, we propose a method to use life log that are automatically collected together with video during an event. Here, data in life log are GPS, pulse and sound. Three kinds of important points are extracted from the three data, and based on the important points, particular parts of video are extracted. According to subjective evaluation experiments, it is revealed that users can easily remember things by watching the extracted video and can remember details of the events including what were not recorded in the video.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In recent years, a life log is getting popular and used to provide specific services for a particular person. One of them is to retrieve memories. Life log helps us to recall events, activities, accidents, etc., but the huge amount of data in life log make it difficult for us to find out what we really want. In this paper, we propose a method to retrieve "a big day" using a feature value that is calculated from GPS data; i.e., the feature is defined as function of visiting frequency of places. According to experiments results, the proposed method can retrieve "a big day" at a rate of 60% and unusual day at a rate of 90%. As subjective evaluations were also carried out from the perspective of effectiveness, efficiency and satisfaction. The results showed that the proposal method has better performance than a conventional method.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In these days, demands for watching out for the safety of elderly and children are extremely increasing. To make the quality of the watching better, we think that living area plays an important role. Therefore, in this paper, we propose an algorithm to estimate living area of a person from his/her accumulated GPS data. The living area is defined by important places and routes. First, taking visiting frequency into account, important places are extracted, then routes are found so as to connect the important places using best-first search. Experiments are carried out for 3 users and evaluated by precision and recall. We confirmed that the proposed algorithm has better performance than a conventional method.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we propose a method to behavior analysis of persons using GPS data to focus on how human have activity. We evaluated whether or not judging a person is active or not from moving routes data using PCA or NMF classification approach. The evaluation results show that the number of important eigenvalues by PCA and approximation error by NMF is effective whether the person is active or not. In addition, we extracted the pattern of moving routes and we evaluated the difference of moving routes extracted by PCA or NMF. As a result, PCA extracted high frequency patterns. On the other hand, NMF extracted not only high frequency patterns but also low frequency patterns.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

Recently, the application and the service which utilize location information from GPS have highly prevailed. In this paper, we developed the system called FLAG which aggregates the variety of Lifelog under location information. FLAG manages location information discriminate between moving and staying. With using FLAG, we visualize categorized location information on the map and the time table. And implement set the function which registers individual name according to users in the staying state. We also link the location information from FLAG to Twitter using the posting time for an example of aggregating various kinds of Lifelog, This function enables Lifelog to show on the map even if the Lifelog has no positional information. For an evaluation of the FLAG system, we created correct data of staying by six users. And we compared accuracies of staying by using two detection methods. As a result, we confirmed that FLAG can be detected high accuracy staying than the original data.

CiNii Article

CiNii Books

researchmap

Other Link： http://id.nii.ac.jp/1001/00102345/

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Recently, the application and the service which utilize location information from GPS have highly prevailed. In this paper, we developed the system called FLAG which aggregates the variety of Lifelog under location information. FLAG manages location information discriminate between moving and staying. With using FLAG, we visualize categorized location information on the map and the time table. And implement set the function which registers individual name according to users in the staying state. We also link the location information from FLAG to Twitter using the posting time for an example of aggregating various kinds of Lifelog, This function enables Lifelog to show on the map even if the Lifelog has no positional information. For an evaluation of the FLAG system, we created correct data of staying by six users. And we compared accuracies of staying by using two detection methods. As a result, we confirmed that FLAG can be detected high accuracy staying than the original data.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this study, we aimed to construct environmental sound database for various sounds as Wisdom of the crowds. For example, considering environmental noise-pollution problems as a kind of the environmental sound, we need to consider not only signature sound, e.g. car noise and railway noise, but also life-space noise, e.g. festivity noise on streets or in parks. However, daily/widely sound collection is difficult to substantiate by few participant. Therefore, we aimed to measure environmental sound covered a vast area by applying crowdsourcing approach. First, we develop a prototype application which is run on the smart device with Android OS, and then we develop a prototype server system for collection and browsing the collected sound data. Then, we calibrated noise level measured by smart devices and carried out a sound collection experiment for validate an accuracy of sensors on the smart devices. In this report, we introduce the environemental sound collection system and the sound-collection experiment using the system.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We discuss a behavior-anaysis method based on strucural features in networks constructed from a personal location history. In this paper, a directed network constructed from GPS location history is called a stay network. Stay network treat set of stay extratcted from the location history as nodes in networks. Generally, nodes of directed networks have out-degrees and in-degrees as structural features. The stay network has biased values of out-degrees, in-degrees and their ratio. Therefore, we assumed existance of relationships between the degree of biased values and human behaviors in the stay, and analyzed the relationships. We forcused on a purpose of the stay, which is assumed to be occur as a result of human behavior, and particularly analyzed relationships between the degree of biased values and the purpose of the stay.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We assess dialog strategies from user's workload standpoint to suggest usage of spoken dialog systems while driving. We evaluate several dialog strategies, which are combinations of methods of dialog initiative and confirmation manner, with objective evaluation method using computer simulation. Two conditions of the simulation are set up; One is that the system speaks if the user has leeway to talk with the system, and the another one is that the system speaks even if the user will be fail to recognize the system's. We also evaluate spoken dialog systems applying these methods with subjective evaluation method. As a result of the evaluations, user initiative strategy has advantages in lower turn number and lower task completion rate than both system initiative strategy and mixed initiative strategies when cognitive rate is high. The result also shows that system initiative strategy and mixed initiative strategy have advantages in lower turn number and lower task completion rate than user initiative strategy when cognitive rate is low. Additionally, the result shows that the method, which system speaks only when user has enough time driving operation, makes user's workload low, however, it need more time to complete tasks.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we tried to extract working patterns by nonnegative matrix factorization using PC operation logs. Experiments were carried out for three occupations, in terms of daily-basis working patterns, we successfully extracted particular patterns for each occupation and some common working patterns for all occupations. We also presented that the extracted patterns can be easily interpreted as the ways of working.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We assess dialog strategies from user's workload standpoint to suggest usage of spoken dialog systems while driving. We evaluate several dialog strategies, which are combinations of methods of dialog initiative and confirmation manner, with objective evaluation method using computer simulation. Two conditions of the simulation are set up; One is that the system speaks if the user has leeway to talk with the system, and the another one is that the system speaks even if the user will be fail to recognize the system's. We also evaluate spoken dialog systems applying these methods with subjective evaluation method. As a result of the evaluations, user initiative strategy has advantages in lower turn number and lower task completion rate than both system initiative strategy and mixed initiative strategies when cognitive rate is high. The result also shows that system initiative strategy and mixed initiative strategy have advantages in lower turn number and lower task completion rate than user initiative strategy when cognitive rate is low. Additionally, the result shows that the method, which system speaks only when user has enough time driving operation, makes user's workload low, however, it need more time to complete tasks.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese  

ライフログとは，人間の行動をデジタルデータとして記録に残すことである．ライフログデータを用いて自分の行動を振り返ることを考える．例えば，現在広く普及しているSocial Networking Service（SNS）の過去の発言や写真を見返すことで，振り返りが可能であるが，それらに記録するのは自分の意志で記録したいと思ったことに限られてしまうという問題点がある．無意識に記録できるライフログデータの一つに，Global Positioning System (GPS) による位置情報データ（GPS データ）があげられる．本研究では，GPSデータから得られる滞在地の特徴量を利用し，振り返りたいと感じる「特別な日」を検索する方式を提案した．評価実験により，提案方式で「特別な日」が高い精度で検索可能であることが示された．

CiNii Article

CiNii Books

researchmap

Language：Japanese  

近年、高齢者による徘徊行動が社会問題となっている。既存の見守りシステムではあらかじめ行動範囲を設定する必要があり、その範囲も詳細に決定できないため十分な見守りが不可能である。 そこで、本研究では蓄積したGPSデータを用いて見守りシステムに応用するための生活圏(行動範囲)を抽出する方法を検討する。GPSデータに含まれる位置情報(緯度経度)をGEOHEXを用いてHEXコードという量子化符号に変換する。それらのHEXコードを滞在地と経路に分類し、頻度の高いHEXコードをそれぞれの生活圏として抽出し、滞在地に属するデータの生活圏と経路に属するデータの生活圏の整合性を評価した。

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Authorship：Lead author   Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this study, we aimed to construct environmental sound database for various sounds as Wisdom of the crowds. For example, considering environmental noise-pollution problems as a kind of the environmental sound, we need to consider not only signature sound, e.g. car noise and railway noise, but also life-space noise, e.g. festivity noise on streets or in parks. However, daily/widely sound collection is difficult to substantiate by few participant. Therefore, we aimed to measure environmental sound covered a vast area by applying crowdsourcing approach. First, we develop a prototype application which is run on the smart device with Android OS, and then we develop a prototype server system for collection and browsing the collected sound data. Then, we calibrated noise level measured by smart devices and carried out a sound collection experiment for validate an accuracy of sensors on the smart devices. As a result of the experiment, we collected nine hundred minutes of sound data, and analyzed the relationships between the measured noise level and some subjective evaluations.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Fuel-consumption meters are equipped with many vehicles, however, they can only show the fuel-consumption/-efficiency value but not allow to use for other purpose, e.g., gathering, analyzing, etc. One of a method to output vehicles data is to use a diagnostic connector having compliant with OBD2 standards, that can output several vehicles' signals, such as, velocity, revolution of engine, and fuel consumption. However, because the protocols depend on manufactures or types of the vehicles, the method is not easy to use for the public. In this study, we aim to estimate the fuel consumption using acoustic signals and several sensor's signals equipped with a smartphone. An estimation of a number of revolutions of engine and an estimation of torque are needed for the estimation of fuel consumption. For the estimation of the number of revolutions, we analyze the acoustic signals from the engine by fast Fourier transform and calculate the estimation value from acoustic signal reducing road-noise approximated as a Gamma mixture distribution. For the estimation of the torque, we use physics of the car with the outputs of several sensors and the vehicle's data. We finally get the fuel consumption refer to a table of fuel-consumption rate, which is created in advance, by the estimated number of revolutions and the estimated torque. As a result of a experiment for the estimation of fuel-consumption, we can achieved a acceptable values of instantaneous fuel consumption, although values of average fuel comsumption have some errors.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we propose a novel technique for behavior analysis focusing on a structure of a personal "stay" network obtained by GPS data. We analyzed network features, which is called scale-free property, small-world property and cluster-state property, for six subjects. We also analyzed a distribution of a feature. that is called "motif, and then compared with other networks based on these features. We evaluated biases of a degree of a hub and a unique number of connection nodes, and clarified a difference between distributions of each subjects' networks.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

When we tell a story of our experience, we try to describe the story replacing our experience with other's one that can recall a similar impression to help understanding of the story. In this paper, we study about an extraction method of a common sense of impressive event between people having different backgounds and experiences. First, we showed several emotional words to the subjects and then asked them to recall the events matching the words. Using the recalled events, we compared two method for evaluate the similarity of the events; a method is based on quantification of the events by his/her emotion with five point scale, and the another one is based on decision the similarity of the event by discussion between two subjects. Experimental result suggested that the decision of the similarity by discussion is heavily affected by the strong emotion of the event.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：Acoustical Society of Japan  

DOI： 10.20697/jasj.69.9_519

CiNii Article

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

On a real environment speech-oriented information guidance system, a valid and invalid input discrimination is important as invalid inputs such as noise, laugh, cough and utterances between users lead to unpredictable system responses. Generally, acoustic features are used for discrimination. Comparing acoustic likelihoods of GMMs (Gaussian Mixture Models) from speech data and noise data is one of the typical methods. In addition to that, using linguistic features is considered to improve discrimination accuracy as it reflects the task-domain of invalid inputs and meaningless recognition re...

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

Recently most individuals have come to use mobile information devices, and daily upload the information obtained by such devices to Internet Cloud. Accordingly the applications of speech information processing have been changing drastically. We need to create a new paradigm for the research and development of speech information processing to adapt to this change. In this paper, we summarize the state-of-the-art speech technologies, propose how to create a research platform for this new paradigm, and discuss the problems we should solve to realize it.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

In this paper we propose a novel speech service software for speech-oriented guidance systems. This software has been developed based on Takemaru-kun system, that is implemented at a community center since Nov. 2002. It is consisted of several modules, such as, Automatic Speech Recognition, Dialog Management, Text-to-Speech, Internet browser, and Computer Graphic Agent. This software and toolkit is plan to be freely distributed. It will be used as the speech service software as Software-as-a-Service (SaaS) for WWW site developers, and also used for an upgrade system of our system for advanc...

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

One of the main applications of Blind Source Separation (BSS) is to improve performance of Automatic Speech Recognition (ASR) systems. However, conventional BSS algorithm has been applied only to speech signals as a pre-processing approach. In this paper, a closely coupled framework between FDICA-based BSS algorithm and speech recognition system is proposed. In the source separation step, a confidence score of the separation accuracy for each frequency bin is first estimated. Subsequently, by employing multi-band speech recognition system, acoustic likelihood is calculated in the Mel-scale filter bank energy using the estimated BSS confidence scores. Therefore, our proposed method can reduce ASR errors which caused by separation errors in BSS and permutation errors in ICA, as in the conventional approach. Experimental results showed that our proposed method improved word correct rate of ASR by 8.2 % and word accuracy rate by 5.7 % on average.

CiNii Article

CiNii Books

researchmap

CiNii Article

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Publisher：電子情報通信学会  

CiNii Article

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

We developed an ambient system that plays a music suitable for the mood of a human-human conversation using words obtained from a continuous speech recognition system. Using the correspondent between a document space based on the texts related to the musics and an acoustic space that express various audio features, the continuous speech recognition results are mapped to an acoustic space. Proper names, which are not coverd by the continuous speech recognizer, are recognized by a wordspotter. In this paper, we show the result of the perfomance evaluation for the system. For read music review texts, the system obtained in MRR of 0.83, which is not bad, with high WER of 70.55%, not low F measure of 31.58. We also show an example result for chat conversations.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

A mathematical model for predicting the user satisfaction of a speech dialogue systems is studied based on a field trial of a voice-navigated music retrieval system. The Subjective Word Accuracy (subjective-WA), of the user is introduced as a background psychometrics for the satisfaction. In the field test, subjective-WA is collected through questionnaires together with satisfactory indexes and various user profiles. First we show that the subjective-WA is more significant to the user satisfactory than (Objective) Word Accuracy (objective-WA ), which is calculated using the manually given transcriptions for the recorded dialogue. Then through top-down clustering of the joint distribution of subjective- and objective-WAs, we show that the user population can be grouped into several sub-groups in terms of sensitivity to recognition accuracy. The lower bound of the objective-WA for the given subjective-WA is also calculated from the joint distribution. Finally, a graphical model is build that predicts the user satisfactory index from user profiles and reduces the distribution uncertainty of user satisfaction by 13% of its variance.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

A mathematical model for predicting the user satisfaction of a speech dialogue systems is studied based on a field trial of a voice-navigated music retrieval system. The Subjective Word Accuracy (subjective-WA), of the user is introduced as a background psychometrics for the satisfaction. In the field test, subjective-WA is collected through questionnaires together with satisfactory indexes and various user profiles. First we show that the subjective-WA is more significant to the user satisfactory than (Objective) Word Accuracy (objective-WA), which is calculated using the manually given transcriptions for the recorded dialogue. Then through top-down clustering of the joint distribution of subjective- and objective-WAs, we show that the user population can be grouped into several sub-groups in terms of sensitivity to recognition accuracy. The lower bound of the objective-WA for the given subjective-WA is also calculated from the joint distribution. Finally, a graphical model is build that predicts the user satisfactory index from user profiles and reduces the distribution uncertainty of user satisfaction by 13% of its variance.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

This paper discusses a training method for the HMM acoustic model that efficiently cover the given vocabulary in order to apply it to the speech interface of a music retrieval system. Customizing the acoustic model to each user is important in this application because 1) song titles and artist names contain many phonetic contexts that are rare in general, e.g. text reading corpora, and 2) the songs stored in a device are different among users. In particular, finding an optimal state-tying structure for the given vocabulary is a new problem in acoustic model training. We propose a method for building a task-dependent acoustic model that uses task-related synthetic utterances of more than one hundred speakers by means of HMM-based speech synthesis. From the experimental evaluation using field test data, we confirmed that the task-dependent acoustic model trained by the proposed method can reduce word error rate by 10% compared to a task-independent model.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

We developed a user customizable speech dialogue system and a framework for automatic speech data collection in field experiments over the Internet. Users can download and install the speech dialogue system onto their own PCs and customize the system on a remote server for their own use. The speech data recorded on their PCs are transferred to the remote server through the Internet. The system enables us to collect speech data spoken by many users with wide variety of acoustic environments. During a two-month field test, we obtained 59 hours of recorded data including 5 hours and 41 minutes detected as speech, which corresponds to 11351 speech segments. The word correct rate for the 4716 speech utterances spoken to the dialogue system was 66.0%, which was improved to 70.5% after applying unsupervised MLLR for each user.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We developed a user customizable speech dialogue system and a framework for automatic speech data collection in field experiments over the Internet. Users can download and install the speech dialogue system onto their own PCs and customize the system on a remote server for their own use. The speech data recorded on their PCs are transferred to the remote server through the Internet. The system enables us to collect speech data spoken by many users with wide variety of acoustic environments. During a two-month field test, we obtained 59 hours of recorded data including 5 hours and 41 minutes detected as speech, which corresponds to 11351 speech segments. The word correct rate for the 4716 speech utterances spoken to the dialogue system was 66.0%, which was improved to 70.5% after applying unsupervised MLLR for each user.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We are developing a music retrieval system for in-car use based on a spoken dialogue interface. The system can retrieve and play musics that a user wants to listen to. We have previously conducted experiments where each subject uses the system for one hour. In the experiments, we have found that the speech recognition performance is improved as the subjects get used to the system, although the degree of training depends on the subject. In this paper, we conduct extended experiments where each subject uses the system over five one-hour sessions. Experimental results for twelve subjects show that the system achieves about 60% relative improvement in recognition performance at the fifth session compared to the first session.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

Recently, various applications equipped with speech recognition are developed. For example, it is used for the car-navigation systems with handsfree operation. There are some systems of music download via the Internet so a music search interface which is easy to use is expected. Then we create a music search system supposing use in the car was used by spoken dialogue. This system can search and play the music what the user want to listen. In this paper, we discuss a detail of the system and spoken dialogue recording with the s)'stem. Experimental results of 150 subjects with a prototype system show that the system could achieve about 80% word correct indoor environment and about 76% word correct in car environment.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This study aims to describe user problems and process of learning skill in using spoken dialogue systems and to reveal how these impact on the evaluation of the system usefulness. For this aim, we designed a new dialogue system and carried out a field test for a large number of subjects and asked them to evaluate the usefulness of the system. The results showed that the evaluation of the system did not correlate a recognition rate but user satisfaction and comprehension. This suggested that the spoken dialogue systems should be evaluated in terms of user factors. Controlled experiments are needed to discuss in detail.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：Information Processing Society of Japan (IPSJ)  

This study aims to describe user problems and process of learning skill in using spoken dialogue systems and to reveal how these impact on the evaluation of the system usefulness. For this aim, we designed a new dialogue system and carried out a field test for a large number of subjects and asked them to evaluate the usefulness of the system. The results showed that the evaluation of the system did not correlate a recognition rate but user satisfaction and comprehension. This suggested that the spoken dialogue systems should be evaluated in terms of user factors. Controlled experiments are needed to discuss in detail.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：Information Processing Society of Japan (IPSJ)  

In order to develop an open source platform of the speech recognition on Personal Digital Assistant (PDA), a general-purpose speech recognition engine Julius/Julian is ported to the PDA environment. From the experimental evaluations the following results are obtained. In die isolated word recognition, 90% accuracy is obtained by about 1.9 times of real time, that in about 73 times to a standard PC environment. In the connected digit recognition, 99% word accuracy is obtained using HMMs trained by sentences recorded by PDA.

CiNii Article

CiNii Books

researchmap

Event date： 2020.12.8 - 2020.12.9

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2020.12.7 - 2020.12.10

Language：English   Presentation type：Oral presentation (general)  

Venue：Online/Virtual Conference (Auckland, New Zealand)  

researchmap

Other Link： https://ieeexplore.ieee.org/document/9306250

Event date： 2020.11.16 - 2020.11.19

Language：English   Presentation type：Oral presentation (general)  

Venue：Online/Virtual Conference (Osaka, Japan)  

researchmap

Event date： 2020.10.25 - 2020.10.29

Language：English   Presentation type：Oral presentation (general)  

Venue：Online/Virtual Conference (Shanghai, China)  

researchmap

Event date： 2020.10.24

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2020.9.9 - 2020.9.11

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：オンライン  

researchmap

Event date： 2020.9.9 - 2020.9.11

Language：Japanese   Presentation type：Poster presentation  

Venue：オンライン  

researchmap

Event date： 2020.6.24 - 2020.6.26

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2020.6.24 - 2020.6.26

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2020.6.6 - 2020.6.7

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2020.6.6 - 2020.6.7

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2020.5.4 - 2020.5.8

Language：English   Presentation type：Oral presentation (general)  

Venue：Online/Virtual Conference (Barcelona, Spain)  

researchmap

Event date： 2020.3.17 - 2020.3.18

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2020.3.16 - 2020.3.18

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2020.3.16 - 2020.3.18

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2020.3.16 - 2020.3.17

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2019.11.30

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2019.11.18 - 2019.11.21

Language：English   Presentation type：Oral presentation (general)  

Venue：Lanzhou, China  

researchmap

Event date： 2019.11.18 - 2019.11.21

Language：English   Presentation type：Oral presentation (general)  

Venue：Lanzhou, China  

researchmap

Event date： 2019.9.4 - 2019.9.6

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2019.8.31 - 2019.9.1

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：広島県福山市  

researchmap

Event date： 2019.8.20 - 2019.8.25

Language：English   Presentation type：Oral presentation (general)  

researchmap

Event date： 2019.7.3 - 2019.7.5

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2019.7.3 - 2019.7.5

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2019.6.22 - 2019.6.23

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2019.6.22 - 2019.6.23

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2019.3.3 - 2019.3.5

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2019.3.3 - 2019.3.5

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2019.3.3 - 2019.3.5

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2018.11.21 - 2018.11.22

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2018.9.19 - 2018.9.21

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2018.9.15 - 2018.9.17

Language：English   Presentation type：Oral presentation (general)  

Venue：岡山大学 津島キャンパス  

researchmap

Event date： 2018.9.12 - 2018.9.14

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2018.9.12 - 2018.9.14

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2018.9.2 - 2018.9.6

Language：English   Presentation type：Poster presentation  

Venue：Hyderabad, India  

researchmap

Event date： 2018.6.28 - 2018.6.29

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2018.3.13 - 2018.3.15

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2018.3.13 - 2018.3.15

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2018.3.13 - 2018.3.15

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2017.12.12 - 2017.12.15

Language：English   Presentation type：Oral presentation (general)  

Venue：Kuala Lumpur, Malaysia  

researchmap

Event date： 2017.12.12 - 2017.12.15

Language：English   Presentation type：Poster presentation  

Venue：Kuala Lumpur, Malaysia  

researchmap

Event date： 2017.11.7 - 2017.11.10

Language：English   Presentation type：Poster presentation  

Venue：Redondo Beach, CA, USA  

researchmap

Event date： 2017.9.25 - 2017.9.27

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2017.9.25 - 2017.9.27

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2017.9.25 - 2017.9.27

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2017.9.11 - 2017.9.15

Language：English   Presentation type：Poster presentation  

Venue：Maui, Hawaii, USA  

researchmap

Event date： 2017.8.20 - 2017.8.24

Language：English   Presentation type：Poster presentation  

Venue：Stockholm, Sweden  

researchmap

Event date： 2017.6.28 - 2017.6.30

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2017.6.22 - 2017.6.23

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2017.3.22 - 2017.3.25

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2017.3.15 - 2017.3.17

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2017.3.15 - 2017.3.17

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.12.13 - 2016.12.16

Language：English   Presentation type：Oral presentation (general)  

Venue：Jeju, Korea  

researchmap

Event date： 2016.11.28 - 2016.12.2

Language：English   Presentation type：Poster presentation  

Venue：Honolulu, Hawaii  

researchmap

Event date： 2016.11.19 - 2016.11.20

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2016.9.14 - 2016.9.16

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2016.9.14 - 2016.9.16

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.9.14 - 2016.9.16

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.9.14 - 2016.9.16

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2016.9.12 - 2016.9.16

Language：English   Presentation type：Poster presentation  

Venue：Heidelberg, Germany  

researchmap

Event date： 2016.9.12 - 2016.9.16

Language：English   Presentation type：Poster presentation  

Venue：Heidelberg, Germany  

researchmap

Event date： 2016.8.24 - 2016.8.25

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.7.11 - 2016.7.15

Language：English   Presentation type：Oral presentation (general)  

Venue：Seattle, WA, USA  

researchmap

Event date： 2016.7.6 - 2016.7.8

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.7.6 - 2016.7.8

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.5.12 - 2016.5.13

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2016.3.9 - 2016.3.11

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2016.1.21 - 2016.1.22

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2015.12.16 - 2015.12.19

Language：English   Presentation type：Oral presentation (general)  

Venue：Hong Kong  

researchmap

Event date： 2015.11.21 - 2015.11.22

Language：Japanese   Presentation type：Poster presentation  

Venue：岡山大学  

researchmap

Event date： 2015.11.21 - 2015.11.22

Language：Japanese   Presentation type：Poster presentation  

Venue：岡山大学  

researchmap

Event date： 2015.9.16 - 2015.9.18

Language：Japanese   Presentation type：Poster presentation  

Venue：会津大学  

researchmap

Event date： 2015.9.16 - 2015.9.18

Language：Japanese   Presentation type：Poster presentation  

Venue：会津大学  

researchmap

Event date： 2015.9.16 - 2015.9.18

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：会津大学  

researchmap

Event date： 2015.9.6 - 2015.9.10

Language：English   Presentation type：Poster presentation  

Venue：Dresden, Germany  

researchmap

Event date： 2015.8.2 - 2015.8.7

Language：English   Presentation type：Poster presentation  

Venue：Los Angels, CA, USA  

researchmap

Event date： 2015.8.2 - 2015.8.7

Language：English   Presentation type：Poster presentation  

Venue：Los Angels, CA, USA  

researchmap

Event date： 2015.7.8 - 2015.7.10

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岩手県 安比高原  

researchmap

Event date： 2015.7.8 - 2015.7.10

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岩手県 安比高原  

researchmap

Event date： 2015.5.14 - 2015.5.15

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：津田塾大学小平キャンパス  

researchmap

Event date： 2015.3.27

Language：English   Presentation type：Oral presentation (general)  

Venue：St. Louis, Missouri, USA  

researchmap

Event date： 2015.3.16 - 2015.3.18

Language：Japanese   Presentation type：Poster presentation  

Venue：日本大学理工学部  

researchmap

Event date： 2015.3.16 - 2015.3.18

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：日本大学理工学部  

researchmap

Event date： 2015.3.16 - 2015.3.18

Language：Japanese   Presentation type：Poster presentation  

Venue：日本大学理工学部  

researchmap

Event date： 2015.3.5 - 2015.3.6

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄科学技術大学院大学  

researchmap

Event date： 2015.3.5 - 2015.3.6

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄科学技術大学院大学  

researchmap

Event date： 2015.3.5 - 2015.3.6

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄科学技術大学院大学  

researchmap

Event date： 2015.2.16 - 2015.2.17

Language：Japanese   Presentation type：Oral presentation (invited, special)  

researchmap

Event date： 2015.1.9 - 2015.1.12

Language：English   Presentation type：Oral presentation (general)  

Venue：Las Vegas, USA  

researchmap

Event date： 2014.12.14

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2014.12.14

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2014.12.9 - 2014.12.12

Language：English   Presentation type：Poster presentation  

Venue：Cambodia  

researchmap

Event date： 2014.9.3 - 2014.9.5

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2014.9.3 - 2014.9.5

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.6.28 - 2014.6.29

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.5.29 - 2014.5.30

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.5.24 - 2014.5.25

Language：Japanese   Presentation type：Poster presentation  

researchmap

Event date： 2014.5.22 - 2014.5.23

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.5.15 - 2014.5.16

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.4.30

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学  

researchmap

Event date： 2014.3.24 - 2014.3.28

Language：English   Presentation type：Poster presentation  

Venue：Budapest, Hungary  

researchmap

Event date： 2014.3.18 - 2014.3.21

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.3.11 - 2014.3.13

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：東京電機大学  

researchmap

Event date： 2014.3.11 - 2014.3.13

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：東京電機大学  

researchmap

Event date： 2014.3.10 - 2014.3.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：日本大学  

researchmap

Event date： 2014.3.7 - 2014.3.8

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2014.1.31 - 2014.2.1

Language：Japanese   Presentation type：Oral presentation (general)  

researchmap

Event date： 2013.11

Language：Japanese   Presentation type：Poster presentation  

Venue：鳥取大学  

researchmap

Event date： 2013.10.19

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学  

researchmap

Event date： 2013.10.19

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学  

researchmap

Event date： 2013.10.19

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学  

researchmap

Event date： 2013.10.19

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学  

researchmap

Event date： 2013.9.25 - 2013.9.27

Language：Japanese   Presentation type：Poster presentation  

researchmap