Research

  1. Education in Acoustics
  2. Linguistics research of phonetics and phonology including acoustic phonetics and its application (Applied Linguistics)
  3. Speech science including speech production and articulatory phonetics, hearing science including speech perception and auditory phonetics, and cognitive science related to speech and sounds
  4. Speech perception in noise/reverberation, intelligibility of speech, and speech signal processing (e.g., speech enhancement)
  5. Speech-related welfare technology and supporting system for impaired people, acoustic analysis of speech and language disorders, and speech production and perception of hearing-impaired and elderly people
  6. Development of speech processing algorithm including a real-time system and sound-related applications
  7. Speaker characteristics of speech
  8. Other acoustics topics

1. Education in Acoustics

Fiscal year 2020 was the final year of the KAKENHI project (18K02988), whose main purpose was developing vocal-tract models that easily explain the science behind human speech production (e.g., explaining speech production mechanisms through applications in pedagogical situations at science museums and educational institutions). We developed an anatomical-type model with a flexible tongue. In 2019, we extended the model and developed two versions by adding two different jaw-opening mechanisms. In 2020, we presented the models for comparison at INTERSPEECH 2020. In 2020, we conducted a study on a flexible-lip model we developed from 2019 through 2020.

For sound sources, we investigated and tested the size of reed materials to simulate voices from children through adults. We also tested different sizes for adult men, adult women, and children. Furthermore, we developed a sound source that can produce stable sounds with an air pump. This is important for sanitary reasons, especially considering COVID-19. Instructions to make a set consisting of a vocal-tract model and a sound source with everyday products were also developed for online hands-on activities.

We released STL files of vocal-tract models for 3D printing through the Acoustic-Phonetics Demonstrations (APD) website. The website was originally designed for e-learning of acoustic phonetics. Due to the COVID-19 pandemic, such online materials are in greater demand than before. Therefore, English versions of related YouTube video clips with vocal-tract models were also released in 2020. Studies on human speech production are also utilized in the supervision of English TV program for children, NHK-E “Eigo de Asobo with Orton,” and in the TV program NHK-G “Gatten!”

Collaboration with science museums and educational institutions is key in this project. In FY 2020, we collaborated with institutions in Korea, Canada, Germany, and India. In Japan, Prof. Arai conducted lectures at the Tokyo Medical and Dental University and the National Rehabilitation Center for Persons with Disabilities. We also started a collaboration with Toyohashi University of Technology and reported experimental results on simulations using the supercomputer “Fugaku.”

Other research projects of this topic include:

  • “Perception of mono-syllable /ma/ produced by vocal-tract model”
  • “Speech and language pathologists and education in acoustics” (KAKENHI 20K03074)
  • “Simulation of artificial reed-type vocal fold” (collaboration with Toyohashi University of Technology)

2. Linguistics research of phonetics and phonology including acoustic phonetics and its application (Applied Linguistics)

Research projects on this topic include:

  • “Study on Chinese learners”
  • “Voiced vs. voiceless plosive sounds” (KAKENHI 17F17006)
  • “Vowel changes in Korean”
  • “Visualizing articulatory organs using ultrasounds”
  • “Japanese pitch accents”
  • “Speech analysis of Korean”
  • “Geminate consonants in Italian”
  • “Speech production and perception of learners of Spanish”
  • “Vowels and consonants of Japanese and English”

3. Speech science including speech production and articulatory phonetics, hearing science including speech perception and auditory phonetics, and cognitive science related to speech and sounds

Research projects on this topic include:

  • “Vocalization and psychology”
  • “Towards easy-to-understand speech for announcements”
  • “Pure tone and its linguistic expression”
  • “Multi-modal speech perception”

4. Speech perception in noise/reverberation, intelligibility of speech, and speech signal processing (e.g., speech enhancement)

Research projects on this topic include:

  • “Effects of noise/reverberation on speech production and perception” (collaboration with the University of Auckland, New Zealand in part)

5. Speech-related welfare technology and supporting system for impaired people, acoustic analysis of speech and language disorders, and speech production and perception of hearing-impaired and elderly people

Research projects on this topic include:

  • “Speech perception by elderly people”
  • “Hearing impairment and visualization”
  • “Acoustic analysis of speech by articulation disorders”
  • “Hearing impairment and music”

6. Development of speech processing algorithm including a real-time system and sound-related applications

Research projects on this topic include:

  • “Speech signal processing and sound masking system” (collaboration with the University of Auckland, New Zealand in part)
  • “Digital pattern playback”
  • “Real-time auralization and physical activities”

7. Speaker characteristics of speech

Research projects on this topic include:

  • “Speaker identification and language” (collaboration with the National Research Institute of Police Science)

8. Other acoustics topics

Research projects on this topic include:

  • “Study on pitch perception by music players”
  • “Pronunciations of German songs”
  • “Drums and rhythm”
  • “Analysis of sounds by frogs”