| dc.contributor.author | Renals, Steve | en |
| dc.contributor.author | Morgan, Nelson | en |
| dc.contributor.author | Cohen, Michael | en |
| dc.contributor.author | Franco, Horacio | en |
| dc.coverage.spatial | 4 | en |
| dc.date.accessioned | 2006-05-18T17:26:16Z | |
| dc.date.available | 2006-05-18T17:26:16Z | |
| dc.date.issued | 1992-03 | |
| dc.identifier.citation | Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on, Volume 1, 23-26 March 1992 Page(s):601 - 604. | |
| dc.identifier.issn | 1520-6149 | |
| dc.identifier.uri | Digital Object Identifier 10.1109/ICASSP.1992.225837 | |
| dc.identifier.uri | http://ieeexplore.ieee.org/ | |
| dc.identifier.uri | http://hdl.handle.net/1842/1131 | |
| dc.description.abstract | The authors have previously demonstrated that feedforward networks can be used to estimate local output probabilities in hidden Markov model (HMM) speech recognition systems (Renals et al., 1991). These connectionist techniques are integrated into the DECIPHER system, with experiments being performed using the speaker-independent DARPA RM database. The results indicate that: connectionist probability estimation can improve performance of a context-independent maximum-likelihood-trained HMM system; performance of the connectionist system is close to what can be achieved using (context-dependent) HMM systems of much higher complexity; and mixing connectionist and maximum-likelihood estimates can improve the performance of the state-of-the-art context-independent HMM system. | en |
| dc.format.extent | 383305 bytes | en |
| dc.format.mimetype | application/pdf | en |
| dc.language.iso | en | |
| dc.publisher | IEEE | en |
| dc.title | Connectionist probability estimation in the DECIPHER speech recognition system | en |
| dc.type | Conference Paper | en |
