Towards Universal Speech Discrete Tokens: A Case Study for ASR and TTS.
Yifan YangFeiyu ShenChenpeng DuZiyang MaKai YuDaniel PoveyXie ChenPublished in: CoRR (2023)
Keyphrases
- text to speech
- automatic speech recognition
- speech recognition
- speech synthesis
- spontaneous speech
- prosodic features
- speech signal
- word error rate
- speech corpus
- sequence prediction
- noisy environments
- spoken words
- hidden markov models
- discrete geometry
- case study
- error rate
- english text
- spoken language
- text to speech synthesis
- test bed
- line segments
- speech recognizer
- conversational speech
- pattern recognition
- speech retrieval
- word processing
- broadcast news
- speech recognizers
- speech sounds
- finite number
- hough transform
- discrete version
- neural network