Abstract
Between the end of the chain of waveform coders, which reaches from linear PCM to adaptive predictive coders (APC), and the class of (especially: predictive) vocoders, a “coding gap” of roughly 32-2.4 kbit/s is shown to actually define “medium-rate” speech coding. The fundamental approaches trying to close the gap are exposed, working either in time or frequency-domain. It is discussed, how their weaknesses may be overcome in advanced predictive or generalized “Filter-Bank” Coding (FBC) systems. The basic ideas and problems of vector quantization (VQ) are reviewed. Combinations of VQ with other schemes are addressed as well as other composite coding systems.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
