Improving End-to-End Single-Channel Multi-Talker Speech Recognition.
Wangyou ZhangXuankai ChangYanmin QianShinji WatanabePublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2020)
Keyphrases
- end to end
- speech recognition
- single channel
- speech signal
- multi channel
- sound source
- speech enhancement
- hidden markov models
- language model
- automatic speech recognition
- pattern recognition
- speech synthesis
- independent component analysis
- neural network
- frequency domain
- speech recognition systems
- background noise
- prior information
- noisy environments
- video sequences
- image processing
- linear prediction
- wavelet decomposition
- signal processing
- computer vision