Abstract
This paper describes a method for detecting syllabic nuclei from English utterances on a frame-by-frame basis using bandpass-filtered acoustic energy measurements. No knowledge of the utterance's phonetic composition is used. In the training phase, phones in English utterances read by a female speaker were assigned rank-ordered sonority values. These sonority values were predicted using multiple linear regression where the predictor variables were bandpass-filtered acoustic energy values at the phone's central region. Results show that (1) syllabic nuclei are identified at over 60 percent accuracy, and (2) speech rate, defined as syllabic nuclei per unit time, is estimated at over 80 percent accuracy.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings of 2002 IEEE Workshop on Speech Synthesis |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 39-42 |
| Number of pages | 4 |
| ISBN (Print) | 0780373952, 9780780373952 |
| DOIs | |
| State | Published - 2002 |
| Event | 2002 IEEE Workshop on Speech Synthesis - Santa Monica, United States Duration: Sep 11 2002 → Sep 13 2002 |
Other
| Other | 2002 IEEE Workshop on Speech Synthesis |
|---|---|
| Country | United States |
| City | Santa Monica |
| Period | 9/11/02 → 9/13/02 |
Keywords
- sonority
- speech energy
- speech rate
- syllabic nuclei
ASJC Scopus subject areas
- Language and Linguistics
- Cultural Studies
Fingerprint Dive into the research topics of 'Automatic detection of syllabic nuclei using acoustic measures'. Together they form a unique fingerprint.
Cite this
Kawai, G., & Van Santen, J. (2002). Automatic detection of syllabic nuclei using acoustic measures. In Proceedings of 2002 IEEE Workshop on Speech Synthesis (pp. 39-42). [1224368] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WSS.2002.1224368