Article details

Research area
Natural language & AI

Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on



Increased robustness against bit errors for distributed speech recognition in wireless environments


In distributed speech recognition, the speech features are com­puted on a mobile device, compressed, and sent over a network to a speech recognition server, where the Viterbi search and hid­den Markov modeling takes place. In this work, we examine some error concealment methods for distributed speech recognition over burst error channels. We consider interpolation and interleaving, and we present a novel use of the stochastic weighted Viterbi recog­nition algorithm to increase robustness against interpolated fea­tures. We examine interleaving at both the frame level and code-book index level. Channel errors are simulated using a Gilbert model, and the performance of our algorithm is compared with other techniques, including the ETSI DSR standard, on a digits task and a large vocabulary task. Coupled with interleaving and in­terpolation, our algorithm can provide accuracy as high as 96.7% on a digit recognition task during an average bit error probabil­ity of 20 1.On the more difficult WSJ task, the accuracy without bit errors is 85.7%. Using our algorithm, we can achieve 82.9% accuracy with an average bit error probability of130.

Read/download now