Permutation Invariant Training Based Single-Channel Multi-Talker Speech Recognition with Music Background.
Yuexi ShangguanJiayi YangPublished in: AIAM (IEEE) (2019)
Keyphrases
- speech recognition
- single channel
- speech signal
- speech enhancement
- sound source
- noisy environments
- multi channel
- isolated word
- background noise
- hidden markov models
- acoustic models
- language model
- automatic speech recognition
- speech synthesis
- speech recognizer
- audio features
- pattern recognition
- speech recognition systems
- speaker independent
- speaker identification
- independent component analysis
- frequency domain
- machine learning
- multiscale
- brain activity