A unified DNN approach to speaker-dependent simultaneous speech enhancement and speech separation in low SNR environments.
Tian GaoJun DuLi-Rong DaiChin-Hui LeePublished in: Speech Commun. (2017)
Keyphrases
- speech enhancement
- low snr
- speaker dependent
- speaker identification
- noisy environments
- speech signal
- speech recognition
- phoneme recognition
- speaker independent
- noise reduction
- single channel
- vocal tract
- signal to noise ratio
- automatic speech recognition
- linear prediction
- speaker recognition
- noise model
- sound source
- low signal to noise ratio
- pattern recognition
- speech recognizer
- speaker adaptation
- multi channel
- speaker verification
- spectral analysis
- speech synthesis
- hidden markov models
- similarity measure
- natural images
- edge detection
- wavelet transform