Abstract
In this paper, we extend a previously introduced method for the generation of vocal tract length invariant (VTLI) features. The novelty is a reduction of the number of obtained invariances to the more desired ones, which results in a significant improvement of recognition rates. In experiments on the TIMIT database, the enhanced discrimination capabilities and robustness to mismatches between training and test conditions are shown.
Originalsprache | Englisch |
---|---|
Titel | INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP |
Seitenumfang | 4 |
Band | 3 |
Herausgeber (Verlag) | International Speech Communication Association (ISCA) |
Erscheinungsdatum | 01.01.2006 |
Seiten | 1499-1502 |
ISBN (Print) | 978-160423449-7 |
Publikationsstatus | Veröffentlicht - 01.01.2006 |
Veranstaltung | INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing - Pittsburgh, USA / Vereinigte Staaten Dauer: 17.09.2006 → 21.09.2006 Konferenznummer: 72042 |