Noise Robust Features Based on MVA Post-processing
| Autor: | Rafik Djemili, Djemil Messadeg, Houcine Bourouba, Mohamed Cherif Amara Korba |
|---|---|
| Přispěvatelé: | Mohamed Cherif Messaadia University - Université Mohamed-Chérif Messaadia [Souk Ahras], Badji Mokhtar-Annaba University, University 8 mai 1945, Université 20 Août 1955 Skikda, Abdelmalek Amine, Ladjel Bellatreche, Zakaria Elberrichi, Erich J. Neuhold, Robert Wrembel, TC 5 |
| Jazyk: | angličtina |
| Rok vydání: | 2015 |
| Předmět: |
Computer science
Speech recognition Perceptual linear prediction Subtraction Normalization (image processing) 020206 networking & telecommunications 02 engineering and technology 01 natural sciences ComputingMethodologies_PATTERNRECOGNITION Autoregressive model Moving average 0103 physical sciences 0202 electrical engineering electronic engineering information engineering [INFO]Computer Science [cs] Mel-frequency cepstrum 010301 acoustics |
| Zdroj: | IFIP Advances in Information and Communication Technology 5th International Conference on Computer Science and Its Applications (CIIA) 5th International Conference on Computer Science and Its Applications (CIIA), May 2015, Saida, Algeria. pp.155-166, ⟨10.1007/978-3-319-19578-0_13⟩ IFIP Advances in Information and Communication Technology ISBN: 9783319195773 CIIA |
| DOI: | 10.1007/978-3-319-19578-0_13⟩ |
| Popis: | Part 6: Information Technology: Text and Speech Processing; International audience; In this paper we present effective technique to improve the performance of the automatic speech recognition (ASR) system. This technique consisting mean subtraction, variance normalization and application of temporal auto regression moving average (ARMA) filtering. This technique is called MVA. We applied MVA as post-processing stage to Mel frequency cespstral coefficients (MFCC) features and Perceptual Linear Prediction (RASTA-PLP) features, to improve automatic speech recognition (ASR) system.We evaluate MVA post-processing scheme with aurora 2 database, in presence of various additive noise (subway, babble because, exhibition hall, restaurant, street, airport, train station). Experimental results demonstrate that our method provides substantial improvements in recognition accuracy for speech in the clean training case. We have completed study by comparing MFCC and RSTA-PLP After MVA post processing. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|