Arai Lab. Newsletter 2005

1. Dr. Ben Gold

Dr. Ben Gold, who specialized in the research field of digital signal processing for long time at the Lincoln Laboratory of the Massachusetts Institute of Technology (MIT), passed away on January 15, 2005. I shall miss him deeply. He was the father of digital signal processing, the first person in the world to systematize the field of digital signal processing, and published the first textbook on the subject “Digital Processing of Signals” in 1969.

Digital Processing of Signals published by Dr. Ben Gold in 1969 (left)
and his autograph (right)

After retiring from MIT Lincoln Lab., he gave a series of lectures on speech signal processing at UC Berkeley every two years. When I was in Berkeley in 1997, I had a chance to attend his lectures, and because his office was also located in our institute, International Computer Science Institute (ICSI), this allowed me to meet with him personally almost every week. Owing to this background, I have been very close with Dr. and Mrs. Gold ever since.

At MIT, I was a Visiting Scientist of the Speech Communication Group of Research Laboratory of Electronics (RLE) and conducted research activities with Prof. Kenneth N. Stevens, the head of the group. This collaboration was also achieved because of Dr. Gold as he was a good friend of Prof. Stevens. Before I left MIT in 2004, I had the last chance to see Dr. Gold and showed him my “vocal-tract models” [2001_06] that I had not had the chance to show him with decent time. Dr. Gold delightfully commented “please do come to my lecture at Berkeley next time and show this to my students. Regarding Dr. Gold’s request that I show my model to the students at Berkeley, we talked about the possibilities over and over again. At one time, it was suggested that he show them a video, but eventually I promised him that I would go to Berkeley and give a lecture and demo and we had been making specific arrangements. However, sadly and regretfully Dr. Gold passed away suddenly in January, 2005.

2. Visit to the U.S. in March

Unfortunately, the lecture at Berkeley became the last promise with Dr. Gold. To fulfill our promise, I visited the University of California, Berkeley (UC Berkeley) in March 2005, where Dr. Nelson Morgan (Director of International Computer Science Institute, a UC Berkeley affiliated research organization) kindly allotted me some time out of his lecture “Audio Signal Processing in Humans and Machines” so that I could give a lecture and demo using my “vocal-tract models.” This lecture was originally to be given jointly by Dr. Morgan and Dr. Gold.

Scene from lecture at Berkeley

After Berkeley, I went on to Boston where I met Mrs. Gold and visited MIT. At MIT, I met with the Speech Communication Group and discussedour research with Prof. Stevens and other members. Prof. Stevens welcomed my visit very warmly, and even offered me some time out of his lecture for me to talk to his students. Once again, I gave a lecture and a demo using my “vocal-tract models.” The group also held a welcome lunch for me.

I also visited Dr. Takahashi at CSAIL (Computer Science and Artificial Intelligence Laboratory) at MIT, where we conducted experiments on linking Arai Lab. and MIT by TV Conference for the research described below in Section 7.10. Given that Dr. Takahashi had visited Arai Lab. in February 18, our discussion went very smoothly.

3. Activities for Societies and Awards

3.1 Special Issue at the Journal of the Acoustical Society of Japan

I helped Prof. Okawa of the Chiba Institute of Technology to plan a special issue on “Past, Present, and Future of Phonetics and Phonology” and collected review papers from six authors, namely Kenneth N. Stevens (MIT), John J. Ohala (UC Berkeley), Miyoko Sugito, Kikuo Maekawa, Haruo Kubozono, and Makoto Kuno. Of these papers, the original review paper by Prof. Stevens was carried in the English journal by the Acoustical Society of Japan, “Acoustical Science and Technology.” My Japanese translation of this paper was placed in the special issue of the Japanese journal [2005_04].

3.2 Joint Technical Committee Meeting on “Speech” and “Psychological and Physiological Acoustics”

The technical committee meeting was held jointly by the Technical Committee on Speech and the Technical Committee on Psychological and Physiological Acoustics during the two days on March 3 and 4 at Sophia University, for which I served as the host.  There were 21 presentations in total, of which five were related to the Arai Lab. A tour of three research laboratories related to speech and hearing at Sophia was also held in conjunction.

3.3 PSJ Phonetics Seminar

On March 26, a tutorial seminar based on the theme Acoustic Phonetics was held at Sophia University sponsored by the Phonetic Society of Japan (PSJ), and I was the lecturer of the day. About 170 attended, half of whom were society members and the remaining half non-members. Referring to Prof. Stevens’ lecture back in MIT, I talked about the textbook I am currently writing, while focusing the lecture on demos to the max. The three hours went by very quickly. For those who were not able to attend the lecture, a DVD of the lecture is available.

For details, visit Acoustic_Phonetics_DVD.pdf

Scene from PSJ Phonetics Seminar (left), DVD of the lecture (center), and its booklet (right)

3.4 ASJ Summer Seminar

I also participated in the Acoustical Society of Japan (ASJ) Summer Seminar in Hakuba in summer, where I was in charge of the fundamentals of speech. As a mini activity, I built vocal tract models with the participants using clay. Yuki Nakata, One of my students from the Arai Lab. reported on this summer seminar in the society journal (Journal of the Acoustical Society of Japan, Vol. 61, No. 12, pp. 734-735, 2005).

Scenes of my lecture from ASJ Summer Seminar (left) and the mini activity (right)

3.5 PSJ General Meeting

In the symposium based on the theme “Understanding Phonetics Visually,” I participated as a symposist and talked about “Visualization of speech signals and its phonetical value.” During the presentation, I also discussed the usefulness of vocal-tract models and “Digital Pattern Playback” [2005_26], etc.

3.6 Received the ASJ Poster Award

The poster presentation which I gave during the 2005 Acoustical Society of Japan (ASJ) Spring Meeting received the poster award [2005_20]. The award-giving ceremony was held during the 2005 Autumn Meeting.

Poster which won the ASJ poster award (left) and after the award-giving ceremony (right).

Details of this poster presentation are carried out as an Acoustical Letter of Acoustical Science and Technology (Acoustical Science and Technology, Vol. 27, No. 6, 2006).

3.7 Interspeech

I participated in the international conference on speech communication held in Lisbon, Portugal called “Interspeech.” There were four poster presentations from the Arai Lab. (of which one was mine and another was based on joint research with the Phonetics Laboratory at Sophia University). I served as a reviewer of the Scientific Review Committee of this conference.

3.9 ASA-ASJ Joint Meeting

I was appointed the program committee member of the field of Education in Acoustics in the joint meeting to be held in Hawaii from the end of November to December, 2006. I am also a member of the Committee on Education in Acoustics of the Acoustical Society of America (ASA) as well as the chair of the Research Committee on Education in Acoustics of the Acoustical Society of Japan (ASJ). I participated in the meeting of the Committee on Education in Acoustics during the fall meeting of the ASA held in Minneapolis in October 2005 to start serving as the bridge between the U.S. and Japan.

3.10 International Workshop held at Sophia University in 2006

In March 2006, the technical committee meeting is held jointly by the Technical Committee on Speech and the Technical Committee on Psychological and Physiological Acoustics over two days at the Sophia University. On the first day, an international workshop titled “International Workshop on Frontiers in Speech and Hearing Research” is held. We prepare to invite guest speakers to this workshop as well as to plan for the poster session.

4. Education in Acoustics Using Vocal-Tract Models

4.1 New Vocal-Tract Models

We built a lung model using torso, as well as a head-shaped model with moving tongue. We also built the center casting so that the cylindrical vocal-tract model can be made using clay (used this during the Mini Activity in the ASJ Summer Seminar).

4.2 Launch of Vocal-Tract Model Webpage

We have launched a webpage (in both Japanese and English) introducing the vocal-tract model along with tutorials on vowel production. The URL is:

4.3 Vocal-Tract Model Demonstrations

In 2005, we carried out demos of the vocal-tract model at various venues:

  • Prof. Stevens’ lecture at MIT
  • Lecture at UC Berkeley
  • ASJ Summer Seminar
  • Poster Presentation at the ASJ Spring Meeting
  • Oral Presentation at the ASA fall Meeting
  • PSJ Phonetics Seminar
  • PSJ General Meeting
  • Sophia University Graduate School Lecture “Speech, Acoustics, and Auditory Information Processing
    Of course I gave a lecture and demo here. This lecture is an interdisciplinary class, and in 2005, the participants were students from the Sophia University Graduate School of Electrical and Electronics Engineering, Mechanical Engineering, and Linguistics.
  • Sophia University Lecture Series “Multi-Media Information Synousiacs
    Gave lecture including demo.
  • Faculty of Science and Technology Lecture Series “Biological Science of Human
    Gave lecture including demo.
  • Elizabeth University of Music
    As part of the traveling course given by the Sophia University Community College, I gave a lecture on acoustics at the Elizabeth University of Music (Sophia’s sister school) as well as a demo of the vocal-tract model.
Scenes of my lecture and after the lecture at the Elizabeth University of Music
  • Yokohama Kyoritsu Doremus School (Senior High School)
    Gave a lecture and demo as trial class.
Scenes of my lecture at the Yokohama  Kyoritsu Doremus School
  • Sophia University Open Campus
    Gave a lecture and demo as trial class.
Scenes of my lecture from Sophia University Open Campus
  • Tama Rehabilitation Academy 
    Gave a demo in the intensive course on Acoustics II.
Scenes of my lecture of Acoustics II (Tama Rehabilitation Academy)

4.4 TV Appearance

I appeared in “Kagaku Daisuki Doyou-juku,” an NHK science program for children broadcast on January 15, 2005, and explained briefly and demonstrated the mechanism of vowel production.

Scene from “Kagaku Daisuki Doyou-juku”

4.5 Textbook for Elementary School Students

A photograph of the lung model we built was selected for the elementary school textbook of the prep school SAPIX.

Our lung models in the textbook of SAPIX

5. International and Domestic Collaborations

5.1 MIT

I reported the results of the research with Prof. Stevens during my visit to MIT at the ASJ Spring Meeting and the international conference Interspeech [2005_06]. Ms. Yoko Saikachi, a member of the Speech Communication Group visited the Arai Lab.

5.2 National Center for Rehabilitative Auditory Research (NCRAR)

The paper which Akiko Kusumoto, a graduate of the Arai Lab., wrote when she was at NCRAR based on her master’s thesis was carried in the journal Speech Communication [2005_01].

5.3 Norwegian University of Science and Technology (NTNU)

Nao Hodoshima, a student of the Arai Lab. who had been studying at NTNU for one year, returned to Japan in the summer of 2005. She went to study under Prof. Peter Svensson and Prof. Dawn Behne who are good friends of mine through work on the same research fields from long ago. We continue to conduct collaborative research between NTNU and the Arai Lab.

5.4 Ishikawa National College of Technology

We are collaborating with Prof. Noboru Kanedera in Gmac Project (see Section 6.2). The results can be found in [2005_25].

6. Industry-University Cooperation

6.1 NTT Advanced Technology Co. (NTT-AT)

Vocal tract model VTM-10 provided by Arai Lab. continues to receive high reputation. As we started to run out of stock amidst demands, we manufactured an additional 20 sets.

Educational tool of vocal tract model VTM-10:

6.2 Fujiyama, Inc.

We have been carrying out joint project with Fujiyama Inc. (president: Ms. Junko Yoshii) from 2004. Project members from the Arai Lab. include Yuki Fujikashi, Ayako Koga, Kyoko Matsuura, and Xiaofu Chen. Regarding subtitling using speech detection in digital movies, in 2005, we reported on the results of experiments on automatic detection of speech portions with a small amount of background noise [2005_25].

Fujiyama, Inc.:

6.3 TOA Co.

We are carrying out collaborative research with TOA in the area of preprocessing for improving intelligibility of speech, that Arai Lab. has developed for public address systems. We are especially indebted to Dr. Kiyohiro Kurisu for his support.Project members from the Arai Lab. include Nao Hodoshima, Nahoko Hayashi, Yusuke Miyauchi, Yoshiaki Murakami, Yuki Nakata, and Tadashi Kubo. For results of 2005, see [2005_072005_242005_30].

TOA Co.:

6.4 Texas Instruments Japan Ltd.

Sophia University has been receiving support from Texas Instruments since the start of the University Program. They have been letting us use their DSK (DSP Starter’s Kit) for a long time for “Electrical Engineering Experiments IV” of the Department of Electrical and Electronics Engineering. They presented us with the latest DSK in the summer of 2005, which helped rejuvenate the contents of our experiments. Takahito Goto of the Arai Lab. gave a presentation at the DSPS Educators Conference [2005_33]. He has also participated in the company’s internship and acquired an honorable scholarship.

Texas Instruments Japan Ltd.:

7. Integration of Humanities and Science & Technology at Sophia University

7.1 Grants-in-Aid for Scientific Research (A, 16203041)

“Development of Support System for Communication Disorders and Research on Application to Clinical Situation” applied jointly with Prof. Kyoko Iitaka, Prof. Mitsuko Shindo, Ms. Sawako Hirai (Research Center for Speech and Hearing, Sophia University), Dr. Akemi Iida (School of Media Science, Tokyo University of Technology) etc. with Prof. Tsutomu Sugawara (Department of Linguistics, Sophia University) as the representative researcher entered its second year in 2005 (four-year project on the integration of humanities and science & technology). The Arai Lab. in particular conducted various studies on hearing-impaired and/or elderly people on two out of the three main themes “Development of Easy to Listen Amplification Speech Processing in Reverberant Environment for Hearing-Impaired Listeners,” “Development of Speech Processing Algorithms for Hearing Aids and their Application to Practical Situations,” and “Investigation on Relation between Visual and Auditory Information for Communication Disorders” [2005_012005_022005_032005_072005_092005_112005_132005_142005_182005_202005_222005_232005_242005_262005_282005_292005_30]. We are also planning to cooperate actively in the remaining one theme “Development of Communication Support System for Communication Disorders and Application to Clinical Situation,” and Shimpei Kajima of the Arai Lab. participated in speech recordings.

7.2 Grants-in-Aid for Scientific Research (B, 14310132)

The project “Diagnosis of Children with Difficulties in Written Language Learning and Basic and Clinical Studies on Development of Guidance Program” carried out jointly with Prof. Tsutomu Sugawara, Prof. Jun’ichi Kasajima (Department of Linguistics), etc. led by Prof. Kyoko Iitaka as the representative researcher completed in 2005.

7.3 Grants-in-Aid for Scientific Research (C, 15530629)

The project “Analysis of Phonological Processing Ability of Children with Developmental Phonological Disorders and Development of their Remedial Teaching” carried out jointly with Prof. Kyoko Iitaka, with Ms. Sawako Hirai as the representative researcher also completed in 2005. See [2005_16] for some of the results achieved. In continuation to this, a three-year project “Clarifying the Phonological Processing Characteristics of Children with Developmental Phonological Disorders, Development of Remedial Teaching, and Research on Practical Application”is starting from 2006.

7.4 Grants-in-Aid for Scientific Research (C, 17500603)

With myself as the representative researcher, a two-year project “Development and Improvement of Vocal-Tract Model Based on Speech Production Mechanism of Humans Aiming at Educational Tool for Speech Science” started in 2005. See [2005_062005_102005_202005_25] for some of the results achieved.

7.5 Open Research Center Project

In 2005, I joined the “Information Modeling of System Design Process, and Its Sharing and Reuse” project with Prof. Kiyoshi Ito (Department of Mechanical Engineering) as the representative researcher. 2006 is the last year of the project.

7.6 On-Campus Joint Research

The on-campus joint research “Investigation of Language Development Characteristics of Children with Developmental Phonological Disorders, Development of Early Diagnosis Program, and Application to Therapy” with Ms. Sawako Hirai as representative launched in 2003 ended in 2005.

7.7 Department of Linguistics, Phonetics Laboratory

The Sophia Symposium organized by the Phonetics Laboratory and held at Sophia in December 2004 has become a booklet. It also carries a paper on my presentation [2004_25]. We also carried out joint research with Prof. Tsutomu Sugawara and the members of the Phonetics Laboratory [2005_082005_152005_212005_27]. In December 2005, I served on the thesis committee for the defense of the doctoral thesis of Masahiko Komatsu, a graduate of the Phonetics Laboratory and a joint researcher at the Arai Lab. until March 3. (Thesis title was “Acoustic Constituents of Prosodic Typology”).

7.8 Research Center for Speech and Hearing

Collaboration is carried out with Prof. Kyoko Iitaka, Prof. Mitsuko Shindo, Ms. Sawako Hirai, and members of the Research Center for Speech and Hearing. For example, graduates to whom I provided guidance for their master’s thesis gave presentations in 2005 [2005_122005_17]. I was also given the opportunity to serve on the thesis committee for two master’s theses in 2005.

7.9 Department of Psychology, Cognitive Psychology Lab.

Collaboration is carried out with Prof. Chikashi Michimata and the members of the Cognitive Psychology Laboratory. In particular, I asked Prof. Michimata to review the master’s thesis of Takuma Okuda of the Arai Lab. as a thesis committee in 2005.

7.10 Joint Research with Prof. Yasuhisa Tamura (Department of Mechanical Engineering, Chair of Information System)

Keiichi Yasu of the Arai Lab. has been carrying out joint experiments and co-writing papers with Prof. Tamura on “Synchronous e-learning with use of instant messenger.”

8. Contributions and Activities for Societies, Patents, etc.

8.1 Society Activities

8.1.1 Acoustical Society of Japan (ASJ)

I have been serving as a member of the Research Committee on Education in Acoustics since 2003. I was appointed the chair in 2005 to run the committee. Since 2005, I have also become a member of the Editorial Committee Journal Sub-Committee and International Public Relations Committee.

8.1.2 Phonetic Society of Japan (PSJ)

I continue to serve as a councilor and member of the Planning Committee. In 2005, I also helped prepare the DVD of a video on my lecture at the Phonetics Seminar in March (see 3.3 above).

8.2 Research Grants

8.2.1 Moritani Scholarship

We have been awarded a two-year research grant as of 2005.

8.2.2 The Sound Technology Promotion Foundation

We have been awarded the 2005 research grant.

8.3 Books

8.3.1 Translation “Signals and Systems for Speech and Hearing

The third edition, third print was published on March 5, 2005.

8.3.2 Translation “Introduction to Sound

The second edition, second print was published on April 5, 2005.

8.3.3 Translation “A Basic Introduction to Speech Perception

The second edition was published on January 15, 2005. The review by Prof. Imaizumi of the Hiroshima Prefectural University of the four books we have translated in the past centering around this “A Basic Introduction to Speech Perception” was carried in the Journal of the Acoustical Society of Japan (Journal of the Acoustical Society of Japan, Vol. 61, No. 7, p. 416, 2005).

9. News in the Arai Lab.

9.1 Journal Paper

The paper submitted to the journal Speech Communication has been published at last (see 5.2 [2005_01]).

9.2 Contract Assistant

Keiichi Yasu, a graduate of the Arai Lab., has joined us as our contract assistant in 2005.

9.3 JSPS Research Fellow

Nao Hodojima has been picked as a Research Fellow of the Japan Society for the Promotion of Science (DC2). Part of the research has been done at the Norwegian University of Science and Technology (NTNU).

10. Titles of Theses

10.1 Master’s Theses

Takahito Goto
Development of Speech Processing Systems with a Digital Signal Processor

Takuma Okuda
Study on Difference between Left and Right Cerebral Hemisphere in Speech Perception

Fumihiro Tadokoro
A study for real-time speech pre-processing by steady-state suppression in actual environment: Through the hearing experiment for elderly people

Yukari Hatta
Improving Speech Intelligibility of Elderly People by the Steady-State Suppression: Aiming for Application to Hearing Aids