Interpolative variable frame rate transmission of speech features for distributed speech recognition.
Huiqun DengDouglas D. O'ShaughnessyJean-Guy DahanWilliam F. Ganong IIIPublished in: ASRU (2007)
Keyphrases
- speech recognition
- frame rate
- speech recognition systems
- speech synthesis
- speech signal
- cepstral coefficients
- automatic speech recognition
- speech recognizer
- speech processing
- hidden markov models
- pattern recognition
- speech recognition technology
- language model
- feature space
- speech recognizers
- noisy environments
- speaker identification
- feature extraction
- vector quantization
- speaker recognition
- high speed
- speaker dependent
- motion blur
- mel frequency cepstral coefficients
- speaker independent
- speech recognition errors
- speech retrieval
- video quality
- word error rate
- computational complexity
- noisy speech
- video sequences