Unified Autoregressive Modeling for Joint End-to-End Multi-Talker Overlapped Speech Recognition and Speaker Attribute Estimation.
Ryo MasumuraDaiki OkamuraNaoki MakishimaMana IhoriAkihiko TakashimaTomohiro TanakaShota OrihashiPublished in: CoRR (2021)
Keyphrases
- speech recognition
- end to end
- autoregressive
- automatic speech recognition
- language model
- speech signal
- hidden markov models
- non stationary
- speech recognizer
- noisy environments
- speaker identification
- pattern recognition
- speaker independent
- random fields
- speech synthesis
- speaker diarization
- speaker dependent
- speaker adaptation
- speaker verification
- sar images
- speech recognition systems
- parameter estimation
- mobile devices
- information retrieval
- acoustic models
- neural network