Universal speech/audio coding using hybrid ACELP/TCX techniques

This paper presents a hybrid audio coding algorithm integrating an LP-based coding technique and a more general transform coding technique. ACELP is used in LP-based coding mode, whereas algebraic TCX is used in transform coding mode. The algorithm extends previously published work on ACELP/TCX coding in several ways. The frame length is increased to 80 ms, adaptive multi-length sub-frames are used with overlapping windowing, an extended multi-rate algebraic VQ is applied to the TCX spectrum to avoid quantizer saturation, and noise shaping is improved. Results show that the proposed hybrid coder has consistently high performance for both speech and music signals.

[1]  Roch Lefebvre,et al.  A wideband speech and audio codec at 16/24/32 kbit/s using hybrid ACELP/TCX techniques , 1999, 1999 IEEE Workshop on Speech Coding Proceedings. Model, Coders, and Error Criteria (Cat. No.99EX351).

[2]  V. T. Ruoppila,et al.  Combined speech and audio coding by discrimination , 2000, 2000 IEEE Workshop on Speech Coding. Proceedings. Meeting the Challenges of the New Millennium (Cat. No.00EX421).

[3]  Roch Lefebvre,et al.  Low-complexity multi-rate lattice vector quantization with application to wideband TCX speech coding at 32 kbit/s , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[4]  N. J. A. Sloane,et al.  A fast encoding method for lattice codes and quantizers , 1983, IEEE Trans. Inf. Theory.

[5]  Juin-Hwey Chen,et al.  Transform predictive coding of wideband speech signals , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.