Message from the General Chair

Dear fellow scientists:

On behalf of the organizing committee, I warmly welcome you to INTERSPEECH 2010 in Chiba, Japan.

The study of spoken language processing is inherently interdisciplinary, and is comprised of diverse academic areas ranging from psychology, linguistics, physiology, and physics, to medicine, education, and engineering. Spoken language is unique because it changes its shape and form depending on the language, although the mechanism of speech production is largely identical in all human beings.

International interaction among a myriad of scientific disciplines is essential to better understand spoken language processing and advance its technologies. This need prompted Prof. Hiroya Fujisaki and his colleagues to inaugurate ICSLP (International Conference on Spoken Language Processing) in 1990. The term INTERSPEECH became ICSLP's sub-title in 2000 (Beijing, China), and, when ICSLP and EUROSPEECH merged, became their main title. Therefore, this year marks the 20th anniversary of the birth of ICSLP in Japan, and the 10th anniversary of the birth of INTERSPEECH.

During the past two decades, Japan has hosted 2 ICSLPs (1990 in Kobe and 1994 in Yokohama) but not INTERSPEECH. Your hosts are pleased to provide you with an opportunity to exchange views on our common interests.

Spoken language is a core part of civilization. Spoken language processing technologies have advanced tremendously in recent years. Yet many people are denied access to such advances due to physical problems, economical situations, under-resourced languages, and other limitations. It is incumbent on us spoken language scientists to meet this challenge. This is why INTERSPEECH 2010's theme is "Spoken language processing for all – Regardless of age, health conditions, native languages, and environments. " Keynote speeches, special sessions, and tutorials reflect our ideal.

My special thanks are due to 3 keynote speakers. On 27 September immediately following the opening ceremony, Prof. Steve Young, this year’s ISCA medalist, discusses the state-of-the-art and future of statistically based spoken language technologies in his talk titled "Still talking to machines (cognitively speaking)". On 28 September, Prof. Tohru Ifukube will enlighten us on how speech technology enriches the physically challenged population in his keynote titled "Sound-based assistive technology supporting seeing, hearing, and speaking for the disabled and the elderly" . On September 29, Dr. Chiu-yu Tseng will shed light on the role of prosody in human processing of spoken discourse in her keynote titled " Beyond sentence prosody" .

Spoken language translation is a key to overcome language barriers. Japan has been playing a leading role in this area since 1980s. A session titled "Speech translation technology" is scheduled in the morning of 30 September. My special thanks go to our session speakers for accepting our requests.

I would like to express our thanks to everyone who proposed tutorials and special sessions. 9 tutorials in 3 tracks are offered on September 26. I encourage INTERSPEECH 2010 participants, especially junior scientists, to attend these information-packed opportunities. Special sessions that reflect this year's conference theme have been arranged either independently by our program committee or jointly by proposers and our program committee.

I am deeply grateful for everyone who submitted papers, without which INTERSPEECH cannot survive.

Various public institutions and private corporations have generously agreed to provide financial and material support. More than 20 exhibition booths enrich the conference. Their support and participation are gratefully acknowledged.

The Chiba Convention Bureau and International Center supported us from the very beginning of our planning. The Nippon Travel Agency helped us with conference registration. I offer many thanks to both.

It was towards the end of 2004 that we proposed to host INTERSPEECH 2010. Many colleagues have been involved in the past 6 years. I would like to express my sincere appreciation to all of them, particularly to our program chair Dr. Satoshi Nakamura and members of Keikichi Hirosethe local organizing committee, who dedicated their precious time and efforts for the preparation of the conference. I am also indebted to our advisory members, whose timely advice enabled this conference to become reality.
I wish you all a pleasant stay in Makuhari.

Best regards,

Keikichi Hirose
General Chair, INTERSPEECH 2010

Message from the Technical Program Chair

Welcome to INTERSPEECH 2010. INTERSPEECH is the world’s largest and most comprehensive conference on issues surrounding the science and technology of spoken language processing both in humans and in machines. The theme of this year’s conference is "Spoken language processing for all - Regardless of age, health conditions, native languages, and environments".

We solicited solid research works on a broad range of the following areas which offered innovative contributions from the viewpoint of either a methodological or practical application:

  • Speech Perception and Production
  • Linguistics, Phonology, Phonetics
  • Para-/ Non- linguistic Information
  • Language Processing
  • Analysis, Enhancement and Coding of Speech and Audio Signal
  • Speaker and Language Identification
  • Speech and SL Generation, Speech Synthesis
  • ASR - Signal Processing and Acoustic Modeling
  • ASR - Architecture, Search and Linguistic Components
  • ASR - Robustness and Adaptation
  • LVCSR and its Applications
  • Technologies and Systems for New Applications
  • Spoken Dialogue System, Spoken Language Understanding, Speech Translation, and Information Retrieval
  • Application, Evaluation, Standardization, SL Resources
This year we had 1324 papers submitted, of which 779 are included in the program. The final program consists of 40 oral sessions, 40 poster sessions and three keynotes. In addition to the regular sessions, we opened a call for special sessions and selected eight special sessions. The topics of the special sessions vary reflecting the conference theme: Open Vocabulary Spoken Document Retrieval; Compressive Sensing for Speech and Language Processing; Social Signals in Speech; Quality of Experiencing Speech Services; Speech Intelligibility Enhancement for All Ages, Health Conditions and Environments; INTERSPEECH 2010 Paralinguistic Challenge - Age, Gender, and Affect; Models of Speech - In Search of Better Representations; and Fact and Replica of Speech Production.

Before the main program starts on Monday, you can join the tutorials, which cover state-of-the-art research topics ranging over a wide variety of areas: Discriminative Training - Fundamentals and Applications; Foundations of Statistical Machine Translation: Past, Present and Future; Speech and Language Technology for Linguists and other Human Scientists; Conditional Random Fields and Direct Decoding for Speech and Language Processing; Mobile Voice Search; Multilingual Speech Processing - Rapid Language Adaptation Tools and Technologies; Kernel Engineering for Fast and Easy Design of Natural Language Applications; Meeting Recognition; Medical Speech Processing - Pathologies, Treatment Assistance, Clinical Trials.

The challenging task of reviewing submissions was efficiently handled by the Technical Program Committee consisting of the Area and Sub-area Coordinators: Masahiro Araki, Mary Beckman, Jared Bernstein, Alan Black, Nick Campbell, Jianwu Dang, Li Deng, Pascale Fung, Mark J. F. Gales, Jean-Luc Gauvain, Satoshi Imaizumi, Toshio Irino, Akinori Ito, Naoto Iwahashi, Hisashi Kawai, Diane Kewley-Port, Genichiro Kikui, Norihide Kitaoka, Bastiaan Kleijn, Akinobu Lee, Haizhou Li, Kikuo Maekawa, Kazunori Mano, Tomoko Matsui, Michael McTear, Helen Meng, Roberto Pieraccini, Koichi Shinoda, and Tetsuya Takiguchi.

The submissions were divided into 11 main areas, where the Speech Recognition area has additional four sub-areas, and were reviewed by a total of 663 reviewers. To handle a large number of quality submissions, we had to ask a big favor of the reviewers. We would like to thank their great contributions to the review process. We also have to mention that some of them did help us complete outstanding reviews in a very short period. We sincerely appreciate their contributions.

After the review process, the Area Coordinators carefully analyzed the review results and gathered in Kyoto to fix the final program of the conference. We had an intensive discussion during the two-day-meeting and worked out all the paper judgments and session designing. I would like to express my heartfelt gratitude to all Area/Sub-area Coordinators, ISCA Board, and ISCA president for their support and help.

In addition, I would like to express my thanks to Prof. Minoru Tsuzaki, the Technical Program Committee Co-Chair, for all his support in the whole process starting with preparation of Call for Papers. I would also like to thank Ms. Manami Matsuda for all her hard work as a part of the technical program committee’s secretariat.They conducted all the work in a timely and appropriate manner.

LastSatoshi Nakamura but not least, I thank you for all the contributions to the INTERSPEECH 2010 conference: for attending sessions, presenting papers, being session chairs, and performing all those other necessary functions. We once again welcome you to the exciting city of Makuhari with its wide variety of attractions: shopping at big malls, watching baseball games at Chiba Marine Stadium, enjoying nature in Mihama-en and other neighborhood parks and gardens.

Satoshi Nakamura
Technical Program Chair, INTERSPEECH 2010