Name
Abstract | ||
|---|---|---|
This letter assesses an improved equalization transformation for robust speech recognition in noisy environments. The proposal is an evolution of the parametric approximation to Histogram Equalization named PEQ into a two-step algorithm dealing separately with environmental and acoustic mismatch. A first parametric equalization is done to eliminate environmental mismatch. These equalized data are divided into classes, and parametrically re-equalized using class specific references to reduce the acoustic mismatch. Experiments have been conducted for Aurora 2 and Aurora 4 databases. A comparative analysis of the experimental results shows significant benefits for databases with high acoustic variability like Aurora 4. |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1109/LSP.2012.2199485 | IEEE Signal Process. Lett. |
Keywords | Field | DocType |
approximation theory,speech recognition,ASR,Aurora 2 databases,Aurora 4 databases,PEQ,acoustic mismatch reduction,class-based parametric approximation,environmental mismatch,environmental mismatch elimination,histogram equalization,improved equalization transformation,parametric equalization,robust speech recognition,two-step algorithm,Feature compensation,histogram equalization,parametric equalization,probabilistic classes,robust ASR | Histogram,Equalization (audio),Pattern recognition,Computer science,Approximation theory,Speech recognition,Adaptive histogram equalization,Parametric statistics,Artificial intelligence,Histogram equalization,Feature compensation | Journal |
Volume | Issue | ISSN |
19 | 7 | 1070-9908 |
Citations | PageRank | References |
3 | 0.39 | 9 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Luz García | 1 | 63 | 9.48 |
| M. Carmen Benítez Ortúzar | 2 | 4 | 1.41 |
| Ángel de la Torre | 3 | 482 | 34.91 |
| José C. Segura | 4 | 481 | 38.14 |