Abstract: Whether a syllable is perceived as stressed or not and whether the stress is strong or weak are hot issues in speech prosody research and speech recognition. A focus of the stress study is on the investigation of the acoustic factors which contribute to the perception of stress level. This study examined all possible acoustic/physiological cues to stress based on data from Annotated Chinese Speech Corpus and proposed that times of vibration of vocal folds (TVVF) reflects stress level best. It is traditionally held that pitch and duration are the most important acoustic parameters to stress. But for Chinese which is a tone language and features special strong-weak pattern in prosody, these two parameters might not be the best ones to represent stress degree. This paper proposed that TVVF, reflected as the number of wave pulses of the vocalic part of a syllable, is the ideal parameter to stress level. Since number of pulses is the integral of pitch and duration (Pulse=?f(pitch)dt), TVVF can embody the effect of stress on both pitch and duration. The analyses revealed that TVVF is most correlated with the grades of stress. Therefore, it can be a more effective parameter indicating stress level.

Yin Zhigang, “The Research of the Relationship between Perceived Stress Level and Times of Vibration of Vocal Folds” International Journal of Advanced Research in Artificial Intelligence(IJARAI), 1(8), 2012. http://dx.doi.org/10.14569/IJARAI.2012.010801

@article{Zhigang2012,
title = {The Research of the Relationship between Perceived Stress Level and Times of Vibration of Vocal Folds},
journal = {International Journal of Advanced Research in Artificial Intelligence},
doi = {10.14569/IJARAI.2012.010801},
url = {http://dx.doi.org/10.14569/IJARAI.2012.010801},
year = {2012},
publisher = {The Science and Information Organization},
volume = {1},
number = {8},
author = {Yin Zhigang}
}