Abstract | ||
---|---|---|
The rejection of unknown words is important in improving the performance of speech recognition. The anti-keyword model method can reject unknown words with high accuracy in a small vocabulary and specified task. Unfortunately, it is either inconvenient or impossible to apply if words in the vocabulary change frequently. We propose a new method for task independent rejection of unknown words, where a new phoneme confidence measure is used to verify partial utterances. It is used to verify each phoneme while locating candidates. Furthermore, the whole utterance is verified by a phonetic typewriter. This method can improve the accuracy of verification in each phoneme, and improve the speed of candidate search. Tests show that the proposed method improves the recognition rate by 4% compared to the conventional algorithm at equal error rates. Furthermore, a 3% improvement is obtained by training acoustic models with the MCE algorithm |
Year | DOI | Venue |
---|---|---|
1998 | 10.1109/ICASSP.1998.674406 | Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference |
Keywords | Field | DocType |
error statistics,maximum likelihood estimation,speech recognition,MCE algorithm,acoustic model training,anti-keyword model method,error rates,maximum likelihood,out-of-vocabulary words rejection,partial utterances verification,performance,phoneme confidence likelihood,phoneme confidence measure,phonetic typewriter,recognition rate,small vocabulary,speech recognition,task independent rejection,tests,unknown words,verification accuracy | Pattern recognition,Computer science,Word error rate,Utterance,Maximum likelihood,Speech recognition,Natural language processing,Artificial intelligence,Out of vocabulary,Vocabulary | Conference |
Volume | ISSN | ISBN |
1 | 1520-6149 | 0-7803-4428-6 |
Citations | PageRank | References |
8 | 0.60 | 3 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Takatoshi Jitsuhiro | 1 | 8 | 0.60 |
Satoshi Takahashi | 2 | 8 | 0.60 |
Kiyoaki Aikawa | 3 | 186 | 28.87 |