Applied mel-frequency discrete wavelet coefficients and parallel model compensation for noise-robust speech recognition.
Zekeriya TufekciJohn N. GowdySabri GurbuzEric K. PattersonPublished in: Speech Commun. (2006)
Keyphrases
- speech recognition
- wavelet coefficients
- noisy environments
- low frequency
- wavelet transform
- hidden markov models
- probabilistic model
- probability distribution
- generalized gaussian
- computer vision
- discrete wavelet transform
- high frequency
- statistical model
- vector quantization
- wavelet domain
- multiresolution
- automatic speech recognition
- computational complexity
- similarity measure