Convolutive Prediction for Monaural Speech Dereverberation and Noisy-Reverberant Speaker Separation.
Zhong-Qiu WangGordon WichernJonathan Le RouxPublished in: CoRR (2021)
Keyphrases
- speech signal
- speech recognition
- noisy environments
- speaker identification
- automatic speech recognition
- linear prediction
- speaker verification
- vocal tract
- speaker recognition
- audio visual
- automatic speech recognition systems
- speaker dependent
- prediction accuracy
- sound source
- speech synthesis
- prediction model
- prediction error
- mel frequency cepstral coefficients
- speech sounds
- broadcast news
- acoustic features
- prosodic features
- prediction algorithm
- data sets
- multi modal
- language model
- speaker independent
- feature vectors
- synthesized speech
- phoneme recognition
- hidden markov models
- speaker diarization
- speech recognizer
- non stationary
- missing data
- probabilistic neural network
- noisy data