LAB LOGA page on LabLog, a review site (not the lab's official page)
Speech Media Lab
Machine-translated · View original
The lab lets machines listen to people's spoken voices so that they can hold conversations with them.
AboutAbout the lab
Listening to voices and exchanging words
The Speech Media Lab studies how to analyze the way people interact by voice and how to have machines recognize and understand it.
The lab builds systems that transcribe natural spoken language, such as in meetings and lectures, and systems that make sense of environments where many kinds of sounds are mixed together.
The lab also works on separating singing voice from accompaniment and transcribing music into scores from music audio signals.
The lab aims to realize humanoid robots that can converse naturally, including nods and timing as well as words.
Key terms
- Speech recognition
- Technology by which a computer listens to a person's spoken voice and converts it into text
- Sound source separation
- Extracting each of the original sounds from a recording in which many sounds are mixed together
- Nonverbal information
- Cues other than the words themselves, such as tone of voice, timing, and nods
NewsLab news
A master's student was shortlisted for the Best Student Paper Award at Interspeech 2026
Professor Tatsuya Kawahara elected an ISCA Fellow
A student won the FY2025 Student Poster Award of the Speech Research Committee
The android ERICA named IEEE's Robot of the Month
ResearchResearch themes
Much of human knowledge creation and transmission takes place through spoken communication. The lab aims to realize systems that analyze the speech media people exchange, automatically recognize and understand it, and can also take the initiative to engage with people.
*Illustrative imageSpeech recognition of spoken language
The lab automatically recognizes real-world spoken language, such as in talks, lectures, and meetings. It studies methods for extracting information and structure from this speech to produce transcripts of talks and meetings, subtitles, and summaries.
*Illustrative imageUnderstanding acoustic environments
The lab recognizes and understands environments with multiple speakers and with various sounds other than speech. It combines many types of sensors and builds on statistical acoustic signal processing.
*Illustrative imageMusic information processing
The lab studies methods for sound source separation and automatic music transcription from music audio signals. Based on this, it develops systems that separate singing voice from accompaniment so that each can be processed and edited.
*Illustrative imageSpoken dialogue with robots
The lab aims to realize robots (androids) that behave like humans and can interact with them. It studies models and systems for spoken dialogue that integrate verbal and nonverbal information.
MembersFaculty
The members of the Speech Media Lab consist of faculty, researchers, support staff, PhD and master's students, exchange students, research students, and undergraduates.As of 2026
- ProfessorTatsuya Kawahara
- Associate ProfessorKeisuke Imoto
- Assistant ProfessorKoji Inoue
- Program-Specific Associate ProfessorLALA Divesh Kanu
- Program-Specific Assistant ProfessorTaiga Mori
- Program-Specific Assistant ProfessorLI Sheng
VoicesStudent voices
Compiled from answers by students who were in the lab. Ratings and reviews reflect individual students' experiences, not the lab's official views.
Students who take part in an interview may receive a thank-you gift worth ¥1,000, regardless of what they say.
On a 4-point scale, ● marks where students' answers fall
Who is this lab for?
What's great about this lab?
What did you learn after joining?
What's the vibe, and what is the professor like?
Where do graduates go?
RelatedLabs similar to Speech Media Lab
Visual Media LabFaculty of Engineering, School of Informatics and Mathematical ScienceIntelligent information systems supporting human communicati…
Symbol Emergence Systems LabFaculty of Engineering, School of Informatics and Mathematical ScienceSymbol emergence robotics: AI and robots that understand lan…
Human-Robot Interaction LabFaculty of Engineering, School of Informatics and Mathematical ScienceRealization of various robots, including 'human-like robots'…
Learning Machines LabFaculty of Engineering, School of Informatics and Mathematical ScienceLearning-based methods for embodied systems (humans, robots,…The information on this page is compiled by the LabLog team from the lab's public information and interviews. For corrections or removal requests, please contact LabLog, not the lab.
Report an errorAbout the images
The top and research theme images are illustrative and were not taken at the lab. They are cropped from Wikimedia Commons images. Cropped CC BY-SA images are provided under the same CC BY-SA license as the original (version shown on each line).
- Top image: “Radio-frequency-anechoic-chamber-HDR-0a” Adamantios / CC BY-SA 3.0 (cropped)
- Speech recognition of spoken language: “Surface waves” Roger McLassus / CC BY-SA 3.0 (cropped)
- Understanding acoustic environments: “Tyto alba (Scopoli, 1769)” Michael Gäbler / CC BY 3.0 (cropped)
- Music information processing: “Organ interior (organ pipes)” Bartłomiej Bulicz / CC BY-SA 3.0 (cropped)
- Spoken dialogue with robots: “Quori socially interactive robot platform IMG” Mary Mark Ockerbloom / CC BY-SA 4.0 (cropped)
- Related lab (映像メディア): “NadiaPiet -ArchivalImages of AI + AIxDESIGNLimitsof Classification -2560x1404” Nadia Piet & Archival Images of AI + AIxDESIGN / CC BY 4.0 (cropped)
- Related lab (記号創発システム): “Robot de serveis Pudu 2022” Marc Auledas-Noguera / CC BY-SA 4.0 (cropped)
- Related lab (ヒューマンロボットインタラクション): “Telenoid™ R4 (tele-presence robot for communication) by ATR Hiroshi Ishiguro Laboratory, exhibited at Miraikan ( by Franklin Heijnen)” Franklin Heijnen / CC BY-SA 2.0 (cropped)
- Related lab (学習機械): “3d-printed-humanoid-robot-manav-mumbai-techfest-2015” Jasonprost / CC BY-SA 4.0 (cropped)
About LabLog
LabLog is a lab review app for students only. Student members at each university interview students in labs and publish what they say. Student voices can only be read by students who sign up with a university email address. Operated by bestiee Inc.
Learn more about LabLog ↗Search engines and AI may reference the basic lab info and rating summaries on this page and the description of LabLog. Student voices are for student members only. Automated collection by programs (scraping, or input to or training of AI) is prohibited under Article 8 of the Terms of Service. If you need data for research or analysis, please contact us.