Releasing videos: INTERSPEECH 2021

Prof. Takayuki Arai had the following presentation in the speech production session at INTERSPEECH 2021; videos related to the presentation were released.

Session:[Virtual] Speech Production
Title:Vocal-tract models to visualize the airstream of human breath and droplets while producing speech

Visualization of aerosol during speech production using a head-shaped vocal-tract model with a lung model:

Lung pressure: high / Output: mouth
Lung pressure: low / Output: mouth
It becomes clear that how far the aerosol reaches does not depend on the lung pressure.
Lung pressure: high / Output: mouth & nostril
We can see that the two air streams from the mouth and the nostril merged and formed a more directional pattern.
Nonwoven fabric mask: wearing / Output: mouth & nostril
Nonwoven fabric mask: wearing / Output: mouth
The effectiveness of wearing the nonwoven fabric mask was demonstrated.

Visualization of particle velocities:

Visualization of droplets during bilabial plosive sound production:

Measurement using vocal-tract model producing consonants /b/, /m/, /w/, /r/, and /l/
(BMW-RL model).
Measurement using 2020 model with flexible lips.
Measurement using BMW-RL model + flexible lip model.

You can find further details of Prof. Arai’s presentation here.

(Updated on 2021/12/28)

カテゴリー: news