Piecewise-linear transformation-based HMM adaptation for noisy speech
Tóm tắt
This paper proposes a new method using a piecewise-linear transformation for adapting phone HMM to noisy speech. Various noises are clustered according to their acoustic properties and signal-to-noise ratios (SNR), and a noisy speech HMM corresponding to each clustered noise is made. Based on the likelihood maximization criterion, the HMM which best matches the input speech is selected and further adapted using a linear transformation. The proposed method was evaluated by recognizing noisy broadcast-news speech. It was confirmed that the proposed method was effective in recognizing numerically noise-added speech and actual noisy speech by a wide range of speakers under various noise conditions.
Từ khóa
#Piecewise linear techniques #Hidden Markov models #Signal to noise ratio #Speech enhancement #Speech recognition #Additive noise #Computer science #Impedance matching #Broadcasting #Speech analysisTài liệu tham khảo
sugimura, 1982, A method of word-multitemplate extraction based on minimization of distance, Proc Speech Tech Committee Meeting ASJ S, 82 64
furui, 2000, Noise adaptation of HMMs using neural networks, Proc of the ISCA ITRW ASR2000, 160
ohtsuki, 1999, Improvements in Japanese Broadcast News Transcription, Proc DARPA Broadcast News Workshop, 231
10.1006/csla.1995.0010
10.1109/ICASSP.1995.479290
martin, 1993, Recognition of noisy speech by composition of hidden Markov models, Proc EUROSPEECH, 1031
10.1109/ICASSP.1992.225929