Researcher

Sadaoki Furui

Also known as 古井貞熙

Sadaoki Furui was a speech researcher from Tokyo whose work on spectral derivatives, known as delta cepstra, greatly improved the accuracy of speech recognition systems. He spent 1970 to 1997 in NTT's laboratories, with a stay at AT&T Bell Laboratories in 1978 and 1979, and his textbook Digital Speech Processing, Synthesis, and Recognition appeared in twenty-six editions between 1985 and 2001. In 1997, he joined the Tokyo Institute of Technology as a professor of computer science, and from 2001 to 2005 he was president of the International Speech Communication Association. He became an ISCA Fellow in 2008, was president of the Toyota Technological Institute at Chicago from 2013 to 2019, and was accredited as a Person of Cultural Merit by the Japanese government in 2016. He died in 2022, having authored or coauthored more than 1,000 papers and books on speech recognition, artificial intelligence and natural language processing.

Born
1945
Died
2022
Works
723
Citations
8,962
h-index
41

Years link to their sources. Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 1968–1978

    Degrees at the University of Tokyo

    Born in Tokyo, he earns bachelor's, master's and doctoral degrees in mathematical engineering and instrumentation physics from the University of Tokyo in 1968, 1970 and 1978.

  2. 1970–1997

    Researcher at NTT labs

    He begins his career in 1970 as a researcher at NTT's Musashino Electrical Communication Labs in Tokyo. From 1978 to 1979, he is a visiting researcher at AT&T Bell Laboratories. In 1991 he becomes a research fellow, and he directs the NTT Furui Research Lab until 1997.

  3. 1980s

    Delta cepstrum and a textbook

    In the 1980s, he investigates human perception of transient sounds at NTT and provides the first quantifiable measurement of the importance of spectral transition for intelligibility. Adding spectral derivatives, which Bell Labs names Delta Cepstrum, greatly improves the accuracy of speech recognition systems. Twenty-six editions of his textbook Digital Speech Processing, Synthesis, and Recognition appear between 1985 and 2001.

  4. 1997–2011

    Professor at Tokyo Institute of Technology

    In 1997, he joins the Tokyo Institute of Technology as a professor of computer science, and in 2007 he becomes dean of its graduate school of information science and engineering. He oversees a Japanese national project to develop a system for automatic understanding and summarization of spontaneous speech.

  5. 2001–2009

    ISCA and APSIPA presidencies

    From 2001 to 2005, he serves as president of the International Speech Communication Association. In 2009, he and colleagues start the Asia-Pacific Signal and Information Processing Association, and he becomes its first president.

  6. 2008–2016

    Honours from ISCA, IEEE and Japan

    ISCA names him a Fellow in 2008 for his research in speech and speaker recognition and his contributions to ISCA as a Board member and President, and awards him its Medal for Scientific Achievement in 2009. He receives the 2010 IEEE James L. Flanagan Speech and Audio Processing Award. In 2016, the Japanese government accredits him as a Person of Cultural Merit.

  7. 2013–2021

    President of TTIC

    He leaves Tokyo in 2013 to become president of the Toyota Technological Institute at Chicago. Matthew Turk, a computer vision researcher, succeeds him in July 2019, and he chairs the institute's board of trustees until 2021.

  8. 2022

    Death in 2022

    He dies in July 2022, having authored or coauthored more than 1,000 papers and books on speech recognition, artificial intelligence and natural language processing.

Career

  1. 2013–2021Toyota Technological Institute at ChicagoPresident, Chair, Board of Trustees
  2. 2009Asia-Pacific Signal and Information Processing AssociationCo-founder, President
  3. 1997–2011Institute of Science TokyoProfessor of Computer Science, Dean, Graduate School of Information Science and Engineering
  4. 1978–1979Bell LabsVisiting Researcher
  5. 1970–1997NTTResearcher; Research Fellow, Director

Education

  1. 1978University of TokyoPhD
  2. 1970University of TokyoMaster's degree
  3. 1968University of TokyoBachelor's degree

Most-cited works

  1. Cepstral analysis technique for automatic speaker verification1981 · IEEE Transactions on Acoustics Speech and Signal Processing · 1,176 citations
  2. Speaker-independent isolated word recognition using dynamic features of speech spectrum1986 · IEEE Transactions on Acoustics Speech and Signal Processing · 749 citations
  3. Advances in Speech Signal Processing1991 · 432 citations
  4. Spontaneous Speech Corpus of Japanese2000 · Language Resources and Evaluation Conference (LREC) · 325 citations
  5. Recent advances in speaker recognition1997 · Pattern Recognition Letters · 287 citations

Sources and credits

Photo: We have no published photo of Sadaoki Furui yet. Send us a free-licensed photo.

  1. Speech Processing Pioneer Sadaoki Furui Dies at 77 - IEEE Spectrum spectrum.ieee.org. Checked 20 October 2026.
  2. Oral-History:Sadaoki Furui - Engineering and Technology History Wiki ethw.org. Checked 20 October 2026.
  3. Sadaoki Furui - Engineering and Technology History Wiki ethw.org. Checked 20 October 2026.
  4. ISCA - ISCA Fellow Program isca-speech.org. Checked 20 October 2026.
  5. Matthew Turk Named Third President of the Toyota Technological Institute at Chicago – Department of Computer Science computerscience.uchicago.edu. Checked 20 October 2026.
  6. TTIC Board of Trustees ttic.edu. Checked 20 October 2026.