Title :
Can voice quality improve mandarin tone recognition?
Author :
Surendran, Dinoj ; Levow, Gina-Anne
Author_Institution :
Comput. Sci. Dept., Univ. of Chicago, Chicago, IL
fDate :
March 31 2008-April 4 2008
Abstract :
We investigate several measures of voice quality (VQ) to improve tone recognition in Mandarin Chinese. We find that band energy measures such as spectral balance (Sluijter and van Heuven, 1996) work better than measures based on glottal flow estimation and harmonic-formant differences. We also determine a set of bands and measures that improve tone classification accuracy on broadcast news speech to 64.1% from 60.4% when added to a traditional pitch-duration-intensity set of features. Most improvement is for the neutral tone, for which the F score increases from 0.345 to 0.619.
Keywords :
feature extraction; speech processing; speech recognition; Mandarin tone recognition; glottal flow estimation; harmonic-formant differences; spectral balance; voice quality; Acoustic measurements; Acoustic signal detection; Broadcasting; Character recognition; Computer science; Energy measurement; Feature extraction; Particle measurements; Speech processing; Speech recognition; Feature extraction; Speech processing; Speech recognition;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4244-1483-3
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2008.4518575
