Digital waveform coding in sub-bands offers attractive possibilities for economically encoding speech at bit rates in the range of 7.2–16 kbits/sec. The design of sub-band coders involves the consideration of a large number of parameters and “tradeoffs.” The number of sub-bands, the partitioning of sub-bands (and gaps between sub-bands), allocation of bits among sub-bands and compromises between bits/sample and bandwidth must be considered. In addition, constraints are introduced by practical considerations of integer-band sampling, multiplexing of the digitized sub-band signals, and synchronizing and framing of the data. Design guidelines, based on extensive computer simulations and subjective comparisons, are presented for choosing these parameters and a method for synchronous multiplexing of the sub-band data, without buffering, is proposed. Several examples of sub-band coders for transmission rates of 7.2, 9.6, and 16 Kbits/sec are presented and compared against that of ADPCM and ADM coders. Tapes will be played to demonstrate the quality of the sub-band coders.