A Purely End-to-end System for Multi-speaker Speech Recognition.
Hiroshi SekiTakaaki HoriShinji WatanabeJonathan Le RouxJohn R. HersheyPublished in: CoRR (2018)
Keyphrases
- speech recognition
- end to end
- automatic speech recognition
- hidden markov models
- language model
- speaker dependent
- speech signal
- pattern recognition
- speech synthesis
- speaker identification
- speech processing
- speech recognition technology
- speech recognizer
- noisy environments
- congestion control
- speaker diarization
- speech recognition systems
- speaker recognition
- speaker independent
- speaker adaptation
- speech retrieval
- cepstral coefficients
- probabilistic model
- neural network