An efficient low bit-rate compression scheme of acoustic features for distributed speech recognition.
Azzedine TouaziMohamed DebyechePublished in: Comput. Electr. Eng. (2016)
Keyphrases
- speech recognition
- compression scheme
- low bit rate
- image compression
- acoustic features
- speech signal
- automatic speech recognition
- compression algorithm
- compression ratio
- vector quantization
- image quality
- hidden markov models
- data compression
- noisy environments
- video compression
- speaker identification
- wavelet transform
- pattern recognition
- coding method
- reconstructed image
- language model
- broadcast news
- image processing
- subband
- bit rate
- mel frequency cepstral coefficients
- bitstream
- lossless compression
- audio features
- multiscale
- neural network
- non stationary
- multiresolution
- speaker recognition