| Paper: | SP-P14.13 (ICASSP 2003 Paper) | ||
| Session: | Acoustic Modeling: Tone, Prosody, and Features | ||
| Time: | Thursday, May 20, 15:30 - 17:30 | ||
| Presentation: | Poster (ICASSP 2003 Presentation) | ||
| Topic: | Speech Processing: Acoustic Modeling for Speech Recognition | ||
| Title: | A PHONE RECOGNIZER HELPS TO RECOGNIZE WORDS BETTER | ||
| Authors: | Georg Stemmer; Universität Erlangen-Nürnberg | ||
| Viktor Zeissler; Universität Erlangen-Nürnberg | |||
| Christian Hacker; Universität Erlangen-Nürnberg | |||
| Elmar Nöth; Universität Erlangen-Nürnberg | |||
| Heinrich Niemann; Universität Erlangen-Nürnberg | |||
| Abstract: | For most speech recognition systems dynamic features arethe only way to incorporate temporal context into the output distributions of the HMMs. In this paper we propose an efficient method to utilize a large context in the recognition process. State scores of a phone recognizerwhich runs in parallel to the word recognizer are computed.Integrating these scores in the HMMs of theword recognizer makes their output densities context-dependent.The approach is evaluated on a set of spontaneous utterances which have been recorded with our spoken dialogue system. A significant reduction of the word error rate has been achieved. | ||
| Back | |||
Home -||-
Organizing Committee -||-
Technical Committee -||-
Technical Program -||-
Plenaries
Paper Submission -||-
Special Sessions -||-
ITT -||-
Paper Review -||-
Exhibits -||-
Tutorials
Information -||-
Registration -||-
Travel Insurance -||-
Housing -||-
Workshops