This paper describes extensions to trellis-based scalar-vector quantization (TB-SVQ) , as novel techniques for coding sources with memory. These techniques are proposed as efficient solutions for the quantization of the excitation in Code-Excited Linear Predictive (CELP) coding algorithms. A new 24 kbit/s low-delay trellis CELP (LD-TCELP) coder is introduced which achieves high-performance coding of narrowband speech with an algorithmic delay of only 3 ms. This coder utilizes the granular structure of the underlying trellis code and the codebook shaping capability of the SVQ approach to effectively capitalize on the gain of higher dimensional (24-D) vector quantization. We present experimental results which show that the LD-TCELP coder outperforms two toll-quality standardized coders: ITU-T G.728 16 kbit/s LD-CELP and ITU-T G.726 32 kbit/s ADPCM. The proposed LD-TCELP coder can also utilize a low-complexity codebook search approach with a computational complexity that is about 75% of that of ITU-T G.728 LD-CELP.

This article presents a method to reduce the complexity of trellis-based scalar-vector quantizers. The proposed method partitions the reproduction alphabet into equal-area regions and shapes the codebook subject to a rate-reduced signaling of these regions. As the shaping bit rate reduces, the coder complexity can be significantly reduced without sacrificing any rate-distortion performance.

The trellis-based scalar-vector quantizer (TB-SVQ) can achieve the rate-distortion performance bound for memoryless sources. This paper extends the scope of this quantizer to coding of sources with memory. First considered is a simple extension, called the predictive TB-SVQ, which applies a closed-loop predictive coding operation in each survivor path of the Viterbi codebook search algorithm. Although the predictive TB-SVQ outperforms all other known structured fixed-rate vector quantizers, due to practical reasons, it may not approach the rate-distortion limit. A new quantization scheme motivated by the precoding idea of Laroia et al. (1993), called the precoded TB-SVQ, is also considered; the granular gain is realized by the underlying trellis code while the combination of the precoder and the SVQ structure provides the boundary gain. This new quantization scheme is asymptotically optimal and can, in principle, approach the rate-distortion bound for Markov sources.

Describes a speech coder based on the intersymbol interference coded quantizer (ICQ). The ICQ is a structured vector quantizer that can realize both boundary and granular gains for sources with memory. It is the quantization dual of the intersymbol interference coder-a transmission scheme for channels with memory. The authors have studied two different suboptimal ICQ codebook search algorithms for speech coding and find that the performance of the ICQ based speech coder is very good at rates over 13 kbps but degrades rapidly at lower rates.