Bridging the Gaps of Both Modality and Language: Synchronous Bilingual CTC for Speech Translation and Speech Recognition.
Chen XuXiaoqian LiuErfeng HeYuhao ZhangQianqian DongTong XiaoJingbo ZhuDapeng ManWu YangPublished in: ICASSP (2024)
Keyphrases
- speech recognition
- target language
- source language
- parallel corpus
- language resources
- machine translation
- comparable corpora
- machine translation system
- isolated word
- bilingual dictionaries
- automatic speech recognition
- cross language information retrieval
- speech signal
- query translation
- speech synthesis
- parallel corpora
- language model
- hidden markov models
- broadcast news
- statistical machine translation
- speech processing
- speech recognizer
- cross lingual
- multi modal
- language processing
- translation model
- speech recognition systems
- pattern recognition
- speaker identification
- noisy environments
- speech recognition technology
- noisy speech
- word error rate
- multilingual retrieval
- speech recognizers
- cepstral coefficients
- recognition engine
- cross language
- language independent
- speech enhancement
- keyword spotting
- speaker independent
- speaker dependent
- probabilistic model
- natural language
- language modeling
- text to speech
- information extraction
- acoustic models
- speech recognition errors
- speech retrieval
- computer vision
- speaker recognition