Adapting a HMM-based recogniser for noisy speech enhanced by spectral subtraction.
Juan Arturo Nolazco-FloresSteve J. YoungPublished in: EUROSPEECH (1993)
Keyphrases
- noisy speech
- hidden markov models
- speech recognition
- spectral subtraction
- speech signal
- background noise
- speech enhancement
- signal to noise ratio
- speech synthesis
- speaker identification
- automatic speech recognition
- noisy environments
- low contrast
- computer vision
- multi channel
- edge detection
- multiresolution
- image processing