STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translation.
Qingkai FangRong YeLei LiYang FengMingxuan WangPublished in: ACL (1) (2022)
Keyphrases
- text to speech
- speech recognition
- text to speech synthesis
- speech signal
- text input
- audio visual
- speech synthesis
- english text
- lexical features
- spontaneous speech
- text recognition
- multi lingual
- spoken language
- text mining
- keywords
- noisy environments
- machine translation system
- language acquisition
- emotion recognition
- vocal tract
- english words
- automatic speech recognition
- co occurrence
- hidden markov models
- database
- hearing impaired