Hybrid Scalar/Vector Quantization of Mel-Frequency Cepstral Coefficients for Low Bit-Rate Coding of Speech.
Laura E. BoucheronPhillip L. De LeonSteven SandovalPublished in: DCC (2011)
Keyphrases
- vector quantization
- speaker recognition
- mel frequency cepstral coefficients
- low bit rate coding
- image compression
- successive approximation
- speech signal
- linear predictive
- speaker identification
- high frequency
- speech recognition
- gaussian mixture model
- acoustic features
- audio visual
- image processing
- visual information
- wavelet transform
- feature vectors