Unsupervised speaker adaptation of deep neural network based on the combination of speaker codes and singular value decomposition for speech recognition.
Shaofei XueHui JiangLi-Rong DaiQingfeng LiuPublished in: ICASSP (2015)
Keyphrases
- singular value decomposition
- speech recognition
- speaker adaptation
- speaker independent
- speaker dependent
- automatic speech recognition
- speech recognizer
- least squares
- language model
- neural network
- hidden markov models
- pattern recognition
- maximum likelihood
- speech synthesis
- speech signal
- speaker identification
- dimensionality reduction
- speech recognition systems
- speaker diarization
- noisy environments
- principal component analysis
- mixture model
- speaker recognition
- non stationary
- unsupervised learning
- feature selection