Cross-Channel Spectral Subtraction for meeting speech recognition.
Yu NasuKoichi ShinodaSadaoki FuruiPublished in: ICASSP (2011)
Keyphrases
- speech recognition
- spectral subtraction
- cross channel
- speaker diarization
- speech signal
- free riding
- automatic speech recognition
- hidden markov models
- language model
- speech synthesis
- noisy speech
- noisy environments
- single channel
- pattern recognition
- speaker identification
- background noise
- machine learning
- additive noise
- multiscale