Name
Abstract | ||
|---|---|---|
Abstract This paper investigates a new front-end processing that aims at improving the performance of speech recognition in noisy mobile environments. This approach combines features based on conventional Mel-cepstral Coefficients (MFCCs), Line Spectral Frequencies (LSFs) and formant-like (FL) features to constitute robust multivariate feature vectors. The resulting front-end constitutes an alternative to the DSR-XAFE (XAFE: eXtended Audio Front-End) available in GSM mobile communications. Our results showed that for highly noisy speech, using the paradigm that combines these spectral cues leads to a significant improvement in recognition accuracy on the Aurora 2 task. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1007/s10772-009-9025-9 | International Journal of Speech Technology |
Keywords | Field | DocType |
distributed speech recognition · gsm · line spectral frequencies · noisy mobile communications · formant-like features | Front and back ends,GSM,Feature vector,Pattern recognition,Voice activity detection,Computer science,Speech recognition,Artificial intelligence,Mobile telephony | Journal |
Volume | Issue | ISSN |
10 | 4 | 1572-8110 |
Citations | PageRank | References |
7 | 0.62 | 5 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Addou Djamel | 1 | 7 | 1.30 |
| Sid-Ahmed Selouani | 2 | 124 | 33.39 |
| Kaoukeb Kifaya | 3 | 7 | 0.96 |
| malika boudraa | 4 | 12 | 3.56 |
| bachir boudraa | 5 | 12 | 4.79 |