Rather than select a single type of codec optimized for a particular application, the authors propose a generic sound-coding technology that combines a range of different coding algorithms based on the current properties of the audio and the channel constraints. Rate-distortion optimization is a promising method for such a framework. Subjective listening tests show the advantages of this kind of flexibility in such applications as Internet radio, portable storage, and mobile communications. Bit allocation is not fixed but flexible, which allows for adaptation to time-varying inputs and changing channel properties.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.