7th International Conference on Spoken Language Processing
September 16-20, 2002
In this paper, we evaluate the performance of several robust speech recognition algorithms in a noisy automobile environment as characterized by the Finnish SpeechDat-Car ASR task . By applying acoustic feature compensation, model compensation, and speech detection algorithms to this task, a 51% reduction in word error rate (WER) was obtained relative to the ETSI standard ASR front-end. In addition, these same techniques achieved an average 35% WER reduction for clean condition training and multiple condition training on a simulated speech-in-noise task as characterized by the Aurora 2 ASR task . The paper also presents alternatives for how these algorithms can be implemented in a distributed speech recognition framework.
Bibliographic reference. Kim, Hong Kook / Rose, Richard C. (2002): "Algorithms for distributed speech recognition in a noisy automobile environment", In ICSLP-2002, 233-236.