A spatio-temporal speech enhancement scheme for robust speech recognition.
Erik M. VisserManabu OtsukaTe-Won LeePublished in: INTERSPEECH (2002)
Keyphrases
- speech recognition
- noisy environments
- speech enhancement
- speech signal
- noisy speech
- hidden markov models
- spectral subtraction
- background noise
- speech synthesis
- language model
- speaker identification
- automatic speech recognition
- pattern recognition
- noise reduction
- vocal tract
- speech recognition systems
- neural network
- single channel
- signal to noise ratio
- gaussian mixture model
- video sequences