Abstract

Noise-vocoded speech sound (NVSS) is the synthesized speech sound whose frequency information is greatly reduced while the amplitude envelope information remains preserved. In NVSS the fundamental frequency does not exist, and a change in amplitude is perceived as a change in not only loudness but also pitch. Original speech having physically same amplitude is not always identical. The phenomenon might occur also in NVSS. The purpose of this study is to examine whether change in not amplitude but loudness creates pitch change in NVSS. Subjects listened to paired original speech Japanese vowels and judged whether the second vowels were perceived louder or softer than the first ones. And they listened to paired noise-vocoded vowels to evaluate whether pitch of the second sounds rises or falls from the first ones. Results show that changes in loudness seem to cause pitch change in NVSS. After this, loudness perception in NVSS will be investigated. The data will show whether pitch of the second sounds change from the first ones or not when loudness of the first and the second sounds are same. And loudness perception in noise-vocoded vowels will be compared with loudness perception in the original speech.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call