Speaker-independent model-based single channel speech separation.
Mohammad H. RadfarRichard M. DansereauAbolghasem SayadiyanPublished in: Neurocomputing (2008)
Keyphrases
- speaker independent
- single channel
- speech recognition
- sound source
- speech signal
- speech enhancement
- speaker dependent
- speaker identification
- multi channel
- hidden markov models
- speech recognizer
- digit recognition
- speech recognition systems
- noisy environments
- speech synthesis
- acoustic models
- audio features
- independent component analysis
- speaker adaptation
- language model
- automatic speech recognition
- frequency domain
- spoken language
- pattern recognition
- vocal tract
- speaker recognition
- linear prediction
- brain activity
- computer vision
- wavelet decomposition
- broadcast news
- wiener filter
- acoustic features
- prior information
- video data
- maximum likelihood
- markov random field
- information retrieval