A solution to the problem of speech recognition with signals distorted by low-bit rate coders is presented in this paper. A model for the coding-decoding distortion, a HMM compensation method to include this model, and an EM-based adaptation algorithm to estimate this distortion are proposed here. Medium vocabulary continuous-speech speaker-independent recognition experiments with 8 kbps G.729(CS-CELP), 13 kbps RPE-LTP (GSM), 5.3 kbps G723.1, 4.8 kbps FS-1016 and 32 kbps G.726(ADPCM) coders show that the approach described in this paper is able to dramatically reduce the effect of the coding distortion and, in some cases, gives a word accuracy higher than the baseline system with uncoded speech. Finally, the EM estimation algorithm requires only one adapting utterance and the approach described is certainly The evolution and popularity of cellular and TCP/IP networks has created the problem of improving the recognition accuracy for speech distorted by low-bit rate coders. The distortion of coding schemes in speech recognizers is difficult to model and is an open problem that cannot be solved by applying conventional noise cancelling techniques [1] such as spectral subtraction [2], cepstral mean subtraction [3] and RASTA

Several mismatch conditions can be modeled as an additive bias. This bias is considered independent of the observation vectors, although this approximation is not always accurate. In this paper the dependence of the bias on the observation vectors is taken into consideration in the context of compensating the GSM coding distortion in speech recognition. However, the results presented here can easily be generalized to deal with other types of mismatch. The coding-decoding distortion is modeled here as feature-dependent. This model is employed to propose an Expectation-Maximization (EM) estimation algorithm of the codingdecoding distortion that is able to cancel the effect of GSM coder with as few as one adapting utterance. Finally, the feature-dependent adaptation can give word error rate (WER) 26 % lower than the featureindependent model. 1.