INTERSPEECH 2004  ICSLP

A solution to the problem of speech recognition with signals corrupted by additive noise and distorted by lowbit rate coders is presented in this paper. The additive noise and the coding distortion are cancelled according to the following scheme: firstly, the pdf of the clean codeddecoced speech is estimated with an additive noise model; second, the pdf of the clean uncoded signal is also estimated with a coding distortion model; and finally, the HMM is compensated by using the expected value of the observation pdf in the context of the stochastic weighted Viterbi (SWV) algorithm. The approach leads to reductions as high as 50% or 60% in word error rate.
Bibliographic reference. Yoma, Nestor Becerra / Brito, Ivan / Molina, Carlos (2004): "The stochastic weighted viterbi algorithm: a frame work to compensate additive noise and lowbit rate coding distortion", In INTERSPEECH2004, 28212824.