Speech Recognition by Denoising and Dereverberation Based on Spectral Subtraction in a Real Noisy Reverberant Environment.
Kyohei OdaniLongbiao WangAtsuhiko KaiPublished in: INTERSPEECH (2012)
Keyphrases
- speech recognition
- speech signal
- noisy environments
- spectral subtraction
- denoising
- automatic speech recognition
- speaker identification
- hidden markov models
- background noise
- noisy speech
- language model
- speech synthesis
- linear prediction
- speaker verification
- speech enhancement
- pattern recognition
- speech recognizer
- noisy images
- additive noise
- vocal tract
- image denoising
- acoustic features
- natural images
- noise removal
- neural network
- gaussian noise
- wavelet domain
- image processing
- sound source
- non stationary
- speech recognition systems
- face recognition