In this paper, a basic Mandarin broadcast news speech recognition system is constructed using the MATBN database. It considers the acoustic modeling for Mandarin base-syllables, particles, and paralinguistic phenomena. It also considers environment-dependent acoustic modeling for three recording environments: studio anchors, outdoor reporters, and outdoor interviewee. Moreover, it incorporates a bigram language model with adaptation using data in MATBN. Syllable recognition rates of 89.64, 84.42and 61.62% were achieved for the three environments of anchors, reporters and interviewees, respectively.
Cite as: Chen, C.L., Wang, Y.R., Chen, S.H. (2004) A Study on Mandarin Broadcast News Speech Recognition. Proc. International Symposium on Chinese Spoken Language Processing, 257-260
@inproceedings{chen04d_iscslp,
author={C.L. Chen and Y.R. Wang and S.H. Chen},
title={{A Study on Mandarin Broadcast News Speech Recognition}},
year=2004,
booktitle={Proc. International Symposium on Chinese Spoken Language Processing},
pages={257--260}
}