Leveraging Phone Mask Training for Phonetic-Reduction-Robust E2E Uyghur Speech Recognition.
Guodong MaPengfei HuJian KangShen HuangHao HuangPublished in: CoRR (2022)
Keyphrases
- speech recognition
- acoustic models
- noisy environments
- hidden markov models
- wall street journal corpus
- speech synthesis
- speech recognizer
- automatic speech recognition
- language model
- speech signal
- speaker independent
- speech processing
- pattern recognition
- isolated word
- speaker identification
- discriminative training
- speech recognition systems
- speech recognition technology
- broadcast news
- speech understanding
- speaker recognition
- speech retrieval
- speech recognizers
- speaker dependent
- machine learning
- background noise
- training process
- bayesian networks
- computer vision