能否帮忙到学校的电子期刊IEEE下载一篇论文, 谢谢,haveaplan@hotmail.com
Real-time word confidence scoring using local posterior probabilities on tree trellis search
Lee, A.; Shikano, K.; Kawahara, T.
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP apos;04). IEEE International Conference on
Volume 1, Issue , 17-21 May 2004 Page(s): I - 793-6 vol.1
Digital Object Identifier 10.1109/ICASSP.2004.1326105
Summary: Confidence scoring based on word posterior probability is usually performed as a post process of speech recognition decoding, and also needs a large number of word hypotheses to get enough confidence quality. We propose a simple way of computing the word confidence using estimated posterior probability while decoding. At the word expansion of stack decoding search, the local sentence likelihoods that contain heuristic scores of unreached segment are directly used to compute the posterior probabilities. Experimental results showed that, although the likelihoods are not optimal, we can provide slightly better confidence measures compared with N-best lists, while the computation is faster than the 100-best method because no N-best decoding is required.
Real-time word confidence scoring using local posterior probabilities on tree trellis search
Lee, A.; Shikano, K.; Kawahara, T.
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP apos;04). IEEE International Conference on
Volume 1, Issue , 17-21 May 2004 Page(s): I - 793-6 vol.1
Digital Object Identifier 10.1109/ICASSP.2004.1326105
Summary: Confidence scoring based on word posterior probability is usually performed as a post process of speech recognition decoding, and also needs a large number of word hypotheses to get enough confidence quality. We propose a simple way of computing the word confidence using estimated posterior probability while decoding. At the word expansion of stack decoding search, the local sentence likelihoods that contain heuristic scores of unreached segment are directly used to compute the posterior probabilities. Experimental results showed that, although the likelihoods are not optimal, we can provide slightly better confidence measures compared with N-best lists, while the computation is faster than the 100-best method because no N-best decoding is required.
