TranUSR: Phoneme-to-word Transcoder Based Unified Speech Representation Learning for Cross-lingual Speech Recognition.
Hongfei XueQijie ShaoPeikun ChenPengcheng GuoLei XieJie LiuPublished in: INTERSPEECH (2023)
Keyphrases
- speech recognition
- speech recognizer
- speech synthesis
- speech recognition systems
- cross lingual
- automatic speech recognition
- hidden markov models
- language model
- speech signal
- pattern recognition
- language modeling
- speaker identification
- speech recognizers
- speaker dependent
- speaker independent
- noisy environments
- translation model
- broadcast news
- data mining
- n gram
- text to speech
- out of vocabulary
- spoken language
- statistical machine translation
- word segmentation
- speaker adaptation
- information retrieval