TY - JOUR
T1 - Finding intonational boundaries using acoustic cues related to the voice source
AU - Choi, Jeung Yoon
AU - Hasegawa-Johnson, Mark
AU - Cole, Jennifer
N1 - Funding Information:
The authors would like to thank Ken Chen for providing the pitch normalization algorithm, Taejin Yoon for help with statistical analysis, and Janet Slifka for valuable discussions. This work was supported by a grant from the University of Illinois Critical Research Initiative.
PY - 2005/10
Y1 - 2005/10
N2 - Acoustic cues related to the voice source, including harmonic structure and spectral tilt, were examined for relevance to prosodic boundary detection. The measurements considered here comprise five categories: duration, pitch, harmonic structure, spectral tilt, and amplitude. Distributions of the measurements and statistical analysis show that the measurements may be used to differentiate between prosodic categories. Detection experiments on the Boston University Radio Speech Corpus show equal error detection rates around 70% for accent and boundary detection, using only the acoustic measurements described, without any lexical or syntactic information. Further investigation of the detection results shows that duration and amplitude measurements, and, to a lesser degree, pitch measurements, are useful for detecting accents, while all voice source measurements except pitch measurements are useful for boundary detection.
AB - Acoustic cues related to the voice source, including harmonic structure and spectral tilt, were examined for relevance to prosodic boundary detection. The measurements considered here comprise five categories: duration, pitch, harmonic structure, spectral tilt, and amplitude. Distributions of the measurements and statistical analysis show that the measurements may be used to differentiate between prosodic categories. Detection experiments on the Boston University Radio Speech Corpus show equal error detection rates around 70% for accent and boundary detection, using only the acoustic measurements described, without any lexical or syntactic information. Further investigation of the detection results shows that duration and amplitude measurements, and, to a lesser degree, pitch measurements, are useful for detecting accents, while all voice source measurements except pitch measurements are useful for boundary detection.
UR - http://www.scopus.com/inward/record.url?scp=26844473312&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=26844473312&partnerID=8YFLogxK
U2 - 10.1121/1.2010288
DO - 10.1121/1.2010288
M3 - Article
C2 - 16266178
AN - SCOPUS:26844473312
SN - 0001-4966
VL - 118
SP - 2579
EP - 2587
JO - Journal of the Acoustical Society of America
JF - Journal of the Acoustical Society of America
IS - 4
ER -