Generalized Weighted-Prediction-Error Dereverberation with Varying Source Priors For Reverberant Speech Recognition.
Toru TaniguchiAswin Shanmugam SubramanianXiaofei WangDung T. TranYuya FujitaShinji WatanabePublished in: WASPAA (2019)
Keyphrases
- speech recognition
- prediction error
- speech signal
- linear prediction
- motion compensated
- hidden markov models
- automatic speech recognition
- language model
- speaker identification
- speech processing
- bit rate
- speech synthesis
- pattern recognition
- speech recognizer
- speech recognition technology
- noisy environments
- spectral analysis
- speech recognizers
- isolated word
- reversible watermarking
- inter frame
- speaker independent
- speaker dependent
- motion vectors
- non stationary
- speech recognition systems
- image quality
- image data
- machine learning
- neural network