Embedding Recurrent Layers with Dual-Path Strategy in a Variant of Convolutional Network for Speaker-Independent Speech Separation.
Xue YangChangchun BaoPublished in: CoRR (2022)
Keyphrases
- speaker independent
- speech recognition
- speaker dependent
- speaker identification
- speech recognizer
- hidden markov models
- convolutional network
- digit recognition
- speech recognition systems
- speech signal
- automatic speech recognition
- acoustic models
- noisy environments
- spoken language
- convolutional neural networks
- speaker adaptation
- broadcast news
- feature extraction
- audio visual
- pattern recognition
- speaker diarization
- gaussian mixture model
- low level
- training data