论文信息 - Improvements in beam search for 10000-word continuous-speech recognition

Improvements in beam search for 10000-word continuous-speech recognition

The authors describe the improvements in a time-synchronous beam search strategy for a 10000-word continuous-speech recognition task. Basically they introduced two measures, namely a tree organization of the pronunciation lexicon and a novel look-ahead technique at the phoneme level. The experimental tests performed showed that the number of state hypotheses could be reduced from 50000 to 3000, i.e., by a factor of about 17. At the same time, the word error rate did not increase. >

Hermann Ney | Reinhold Häb-Umbach | H. Ney | Reinhold Häb-Umbach

[1] Douglas B. Paul,et al. Algorithms for an Optimal A* Search and Linearizing the Search in the Stack Decoder* , 1991, HLT.

[2] Xavier L. Aubert. Fast look-ahead pruning strategies in continuous speech recognition , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[3] Hermann Ney,et al. Continuous mixture densities and linear discriminant analysis for improved context-dependent acoustic models , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[4] Hermann Ney,et al. The use of a one-stage dynamic programming algorithm for connected word recognition , 1984 .

[5] Hermann Ney,et al. A look-ahead search technique for large vocabulary continuous speech recognition , 1991, EUROSPEECH.

[6] Lalit R. Bahl,et al. A Maximum Likelihood Approach to Continuous Speech Recognition , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7] Hermann Ney,et al. Data driven search organization for continuous speech recognition , 1992, IEEE Trans. Signal Process..

[8] Lalit R. Bahl,et al. A fast approximate acoustic match for large vocabulary speech recognition , 1989, IEEE Trans. Speech Audio Process..