Cepstral behaviour due to additive noise and a compensation scheme for noisy speech recognition

The speech cepstral coefficients affected by additive noises are investigat ed. The cepstral vector changes as the level of additive noise increases. T he behaviour of cepstral vector change shows that the cepstral vector shrin ks in its norm and converges to the cepstral vector of the noise. This nonl inear behaviour of the cepstral vector can be approximated by a simple line ar expression. Based on this representation, a model adaptation method is d eveloped using deviation vectors. For every model state mean, a deviation v ector is calculated according to the extracted noise spectrum and a pre-def ined noise-to-signal ratio. During the pattern matching, an optimal scaling factor for the deviation vector is determined frame by frame, and the scal ed deviation vector is added to the state mean of speech models so that the clean speech models are adapted to the noisy environment. Experimental res ults show that the proposed method is effective for white noise and colour noises. It also outperforms the weighted projection measure method in exper iments.