Joint encoding of the waveform and speech recognition features using a transform codec.
Xing FanMichael L. SeltzerJasha DroppoHenrique S. MalvarAlex AceroPublished in: ICASSP (2011)
Keyphrases
- speech recognition
- speech recognition systems
- language model
- speech synthesis
- hidden markov models
- pattern recognition
- cepstral coefficients
- speaker identification
- automatic speech recognition
- speech signal
- feature vectors
- feature extraction
- speech recognizer
- noisy environments
- extracting features
- frequency domain
- speech processing
- feature set
- speech retrieval
- speech recognition technology
- spoken language
- natural language processing
- low level
- machine learning
- speaker dependent
- speech recognizers