Alignment-based codeword-dependent cepstral normalization

This paper proposes the alignment-based codeword dependent cepstral normalization algorithm (ACDC_N) which aims to alleviate the acoustical mismatch that occurs when the speech recognizer faces environmental conditions not observed in the training data. ACDC_N is based on the linear channel model of the environment originally proposed by Acero (1990) and on the CDCN solution to this model. ACDC_N substitutes the codebook (Gaussian mixture model) employed by CDCN for the state distributions employed by the recognizer´s HMMs under the assumption that these HMM distributions will model the associated speech segments better than the general GMM distribution. The feature-frame to HMM-state association is obtained through an alignment of a first decoding-pass hypothesis. From this alignment, ACDC_N obtains an estimate of the environmental parameters (noise and channel vectors) which are then employed to obtain an MMSE estimate of the clean speech vectors, in a way similar to Aero´s method. ACDC_N produces an overall reduction of the error rate of over 30 % in the noise range of 0 to 20 dB on experiments conducted on the Aurora-2 noisy digits database.