APVQ encoder applied to wideband speech coding
Document typeConference report
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Rights accessOpen Access
The paper describes a coding scheme for broadband speech (sampling frequency 16 KHz). The authors present a wideband speech encoder called APVQ (adaptive predictive vector quantization). It combines subband coding, vector quantization and adaptive prediction. The speech signal is split into 16 subbands by means of a QMF filter bank and so every subband is 500 Hz wide. This APVQ encoder can be seen as a vectorial extension of a conventional ADPCM encoder. In this scheme, signal vector is formed with one sample of the normalized prediction error signal coming from different subbands and then it is vector quantized. The prediction error signal is normalized by its gain and normalized prediction error signal is the input of the VQ and therefore an adaptive gain-shape VQ is considered. This APVQ encoder combines the advantages of scalar prediction and those of vector quantization. They evaluate wideband speech coding in the range from 1.5 to 2 bits/sample, that leads to a coding rate from 24 to 32 kbps.
CitationSalavedra, J., Masgrau, E. APVQ encoder applied to wideband speech coding. A: International Conference on Spoken Language Processing. "ICSLP 96: Fourth International Conference on Spoken Language Processing: October 3-6, 1996: Wyndham Franklin Plaza Hotel, Philadelphia, PA, USA: proceedings". Philadelphia, PA: Institute of Electrical and Electronics Engineers (IEEE), 1996, p. 941-944.
|APVQ encoder applied to wideband speech coding.pdf||425,5Kb||View/