On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions
Limiting the decrease in performance due to acoustic environment changes remains a major challenge for continuous speech recognition (CSR) systems. We propose a novel approach which combines the Karhunen-Loève transform (KLT) in the mel-frequency domain with a genetic algorithm (GA) to enhance th...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2003-07-01
|
Series: | EURASIP Journal on Advances in Signal Processing |
Subjects: | |
Online Access: | http://dx.doi.org/10.1155/S1110865703302070 |
id |
doaj-1624c8ee493a4883867d49ba38f6798e |
---|---|
record_format |
Article |
spelling |
doaj-1624c8ee493a4883867d49ba38f6798e2020-11-25T01:41:36ZengSpringerOpenEURASIP Journal on Advances in Signal Processing1687-61721687-61802003-07-012003881482310.1155/S1687617203302070On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse ConditionsSid-Ahmed SelouaniDouglas O'ShaughnessyLimiting the decrease in performance due to acoustic environment changes remains a major challenge for continuous speech recognition (CSR) systems. We propose a novel approach which combines the Karhunen-Loève transform (KLT) in the mel-frequency domain with a genetic algorithm (GA) to enhance the data representing corrupted speech. The idea consists of projecting noisy speech parameters onto the space generated by the genetically optimized principal axis issued from the KLT. The enhanced parameters increase the recognition rate for highly interfering noise environments. The proposed hybrid technique, when included in the front-end of an HTK-based CSR system, outperforms that of the conventional recognition process in severe interfering car noise environments for a wide range of signal-to-noise ratios (SNRs) varying from 16 dB to −4 dB. We also showed the effectiveness of the KLT-GA method in recognizing speech subject to telephone channel degradations.http://dx.doi.org/10.1155/S1110865703302070speech recognitiongenetic algorithmsKarhunen-Loève transformhidden Markov modelsrobustness. |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Sid-Ahmed Selouani Douglas O'Shaughnessy |
spellingShingle |
Sid-Ahmed Selouani Douglas O'Shaughnessy On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions EURASIP Journal on Advances in Signal Processing speech recognition genetic algorithms Karhunen-Loève transform hidden Markov models robustness. |
author_facet |
Sid-Ahmed Selouani Douglas O'Shaughnessy |
author_sort |
Sid-Ahmed Selouani |
title |
On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions |
title_short |
On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions |
title_full |
On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions |
title_fullStr |
On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions |
title_full_unstemmed |
On the Use of Evolutionary Algorithms to Improve the Robustness of Continuous Speech Recognition Systems in Adverse Conditions |
title_sort |
on the use of evolutionary algorithms to improve the robustness of continuous speech recognition systems in adverse conditions |
publisher |
SpringerOpen |
series |
EURASIP Journal on Advances in Signal Processing |
issn |
1687-6172 1687-6180 |
publishDate |
2003-07-01 |
description |
Limiting the decrease in performance due to acoustic environment changes remains a major challenge for continuous speech recognition (CSR) systems. We propose a novel approach which combines the Karhunen-Loève transform (KLT) in the mel-frequency domain with a genetic algorithm (GA) to enhance the data representing corrupted speech. The idea consists of projecting noisy speech parameters onto the space generated by the genetically optimized principal axis issued from the KLT. The enhanced parameters increase the recognition rate for highly interfering noise environments. The proposed hybrid technique, when included in the front-end of an HTK-based CSR system, outperforms that of the conventional recognition process in severe interfering car noise environments for a wide range of signal-to-noise ratios (SNRs) varying from 16 dB to −4 dB. We also showed the effectiveness of the KLT-GA method in recognizing speech subject to telephone channel degradations. |
topic |
speech recognition genetic algorithms Karhunen-Loève transform hidden Markov models robustness. |
url |
http://dx.doi.org/10.1155/S1110865703302070 |
work_keys_str_mv |
AT sidahmedselouani ontheuseofevolutionaryalgorithmstoimprovetherobustnessofcontinuousspeechrecognitionsystemsinadverseconditions AT douglasoshaughnessy ontheuseofevolutionaryalgorithmstoimprovetherobustnessofcontinuousspeechrecognitionsystemsinadverseconditions |
_version_ |
1725040769552613376 |