Voice source information representation based on the instantaneous frequency and events
-
- KAWAHARA Hideki
- Wakayama University
-
- SAKAKIBARA Ken-Ichi
- Health Sciences University of Hokkaido
-
- YATABE Kohei
- Tokyo University of Agriculture and Technology
Bibliographic Information
- Other Title
-
- 瞬時周波数とイベントに基づく音源情報の表現について
Abstract
<p>Repetitive valving of airflow is the excitation source of voiced sounds. This repetition rate closely correlates with the perceptual attributes called “pitch.” Since this repetition is approximately regular in voiced sounds, a harmonically related sum of sinusoids models voiced sounds. However, excitation intervals deviate from periodicity in the onset and offset of voiced regions. Moreover, the actual valving of the vocal fold is affected by the effects of muscle, mucosal, airflow, and reflecting sound waves; consequently, strict harmonic relations do not hold. In this report, we introduce a structured test signal for frequency-modulating the fundamental frequency of the synthetic voiced sounds to investigate relations between the instantaneous frequency of the fundamental component and the repetition rate of valving events.</p>
Journal
-
- Proceedings of the Technical Committee on Speech Communication
-
Proceedings of the Technical Committee on Speech Communication 4 (1), n/a-, 2024-01-19
The Technical Committee on Speech Communication,the Acoustical Society of Japan
- Tweet
Details 詳細情報について
-
- CRID
- 1390581003356438528
-
- ISSN
- 27582744
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
-
- Abstract License Flag
- Allowed