Speech and Audio Coding for Multimedia Communications

We have seen rapid progress in high-quality compression of telephone speech and wideband speech signals. Linear prediction, subband coding, transform coding, as well as various forms of vector quantiza-tion and entropy coding techniques have been used to design efficient coding algorithms which can achieve substantially more compression than was thought possible only a few years ago. In the case of audio coding with its bandwidth of 20 kHz and more, the concept of perceptual coding has paved the way for significant bit rate reductions. The paper will explain basic approaches to such compressions , with concentration on existing and upcom-ing international standards. As typical signal classes we shall consider telephone speech, wideband speech, and wideband audio signals all of which differ in listener expectation of offered quality. The main motivations for low bit rate coding are outlined as well as basic and network-related requirements. It will become obvious that speech and audio coders must be both source-specific and hearing-specific to perform adequately at low bit rates.