Low Bit Rate Speech Coding Using Lattice Vector Quantization and Time-Scale Modification

Abstract:

This paper presents a low bit rate speech coder based on predictive lattice vector quantization (PLVQ) and time-scale modification (TSM). The coding model of proposed vocoder is built on the MELP, in which bit rate reduction is achieved by taking advantage of PLVQ and TSM techniques. PLVQ is used to encode the speech line spectrum pair (LSP) parameters, which has the advantage of lower implementation complexity than multi-stage vector quantization (MSVQ), moreover, it does not require memory for codebook storage. With our speech data base, PLVQ can save up to 4 bits/frame compared to unstructured codebook MSVQ. TSM can change the speed of speech signal with its perceptual characteristics remained. Through appending TSM as previous and post process, speech coding at bit rate about 1.1 kbps could be easily achieved without modifying the vocoder structure.

• Periodical: Advanced Materials Research (Volumes 383-390)
• Pages: 5111-5116
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.383-390.5111
• Online since: November 2011
• Authors: Qiang Xiao, Liang Chen, Chao Geng
• Keywords: Lattice Vector Quantization, Low Bit Rate Speech Coding, LSP Parameters, Time-Scale Modification

References

[1] L.M. Supplee, R.P. Chon, A. McCree, et al., MELP: the new Federal standard at 2400bps, IEEE Processing of ICASSP, 1997, pp.1591-1594.

[2] T. Wang, K. Koishida, V. Cuperman, et al., A 1200/2400bps coding suite based on MELP, IEEE workshop on speech coding, Tsukuba, Japan, 2002, pp.90-92.

DOI: 10.1109/scw.2002.1215734

[3] W.J. Han, E.K. Kim, Y.H. Oh, Multicodebook split vector quantization of LSF parameters, IEEE Signal Processing Letters, 2002, 9(12): 418-421.

DOI: 10.1109/lsp.2002.806057

[4] F. Lahouti, A.K. Khandani, Reconstruction of multi-stage vector quantized sources over noisy channels-application to MELP codec, IEEE Processing of ICASSP, 2004, pp.613-616.

DOI: 10.1109/icassp.2004.1326901

[5] T.R. Fischer, A pyramid vector quantizer, IEEE Trans. on information theory, 1986, IT-32 (4): 568-583.

DOI: 10.1109/tit.1986.1057198

[6] L.H. Fonteles, M. Antonini, High dimension lattice vector quantizer design for generalized Gaussian distributions, IEEE Processing of ICIP, 2007, pp.185-188.

DOI: 10.1109/icip.2007.4379985

[7] J.H. Conway, N.J.A. Sloane, Fast quantizing and decoding algorithm for lattice quantizers and codes, IEEE Trans. on information theory, 1982, IT-28 (2): 227-232.

DOI: 10.1109/tit.1982.1056484

[8] P. Rault, C. Guillemot, Indexing algorithms for Zn, an, Dn, and D++n lattice vector quantizers, IEEE Trans. on multimedia, 2001, 3 (4): 395-404.

DOI: 10.1109/6046.966111

[9] B. Ninness, S.J. Henriksen, Time-Scale Modification of speech signals, IEEE Trans. signal processing, 2008, 56(4): 1479-1488.

DOI: 10.1109/tsp.2007.909350

[10] W. Verhelst, M. Roelands, An overlap-add technique based on waveform similarity (WSOLA) for high quality time-scale modification of speech, IEEE Processing of ICASSP, 1993: 554-557.

DOI: 10.1109/icassp.1993.319366

[11] ITU-T P. 862, Perceptual evaluation of speech quality (PESQ): An objective method for end-to-end speech quality assessment of narrow-band telephone network and speech codecs, (2001).

DOI: 10.1109/icassp.2001.941023

[12] M. Djamah, D. O'Shaughnessy, An efficient tree-structured codebook design for embedded vector quantization, IEEE Processing of ICASSP, 2010, pp.4686-4689.

DOI: 10.1109/icassp.2010.5495190