Aipnet: Generative Adversarial Pre-Training of Accent-Invariant Networks for End-To-End Speech Recognition.
Yi-Chen ChenZhaojun YangChing-Feng YehMahaveer JainMichael L. SeltzerPublished in: ICASSP (2020)
Keyphrases
- speech recognition
- end to end
- high bandwidth
- wall street journal corpus
- wireless ad hoc networks
- internet protocol
- hidden markov models
- discriminative training
- automatic speech recognition
- isolated word
- language model
- speech processing
- speech synthesis
- differentiated services
- acoustic models
- speech signal
- admission control
- congestion control
- speech recognizer
- training process
- speech recognition systems
- speech recognition technology
- pattern recognition
- noisy environments
- speaker dependent
- speaker independent
- spoken language
- bayesian networks