It didn't take long for me to hear that a discussion of the market viability for HD voice was a bit ahead of the game. Most readers don't know what HD voice is. So here is a primer for HD Voice.

High-definition voice, or wideband voice provides a level of improved clarity and better audio for VoIP communications. Current telephone calls are sampled 8,000 times a second and constrained to a sound spectrum between 200Hz and 3.3KHz. It is then transmitted over the network or PTSN at 64kbps. Now most of us know that bit of information from our years in telephony. However, HD voice doubles the sampling rate and increases the range of the sound spectrum from 50Hz to 7KHz. This improves the resolution of the voice dramatically, reducing the need for callers to repeat themselves and reduces missed information. The bandwidth requirement for HD voice is also reduced to 32Kbps or half that of traditional telephony.

Additionally, the human voice has important content beyond 14 kHz and the human ear can be sensitive to 20 kHz. With the exception of the telephone, we experience that full range of sound throughout our daily activities.

The following represents the kHz required or supported by various codecs. In most cases the bit rate needed to support wideband voice requirements is much less than the current 64 kbps.

BW (kHz)                                 Typical bit rate (kbps)

3.3                                           8 (G.729), 56 (G.711)

7                                              10 (G.722.2), 24 (G.722.1), 64 (G.722)

14                                            32 (G.722.1C)

20                                            32 (G.719), 64 (AAC-LD)

22                                            32 (Siren22)

All of the above is interesting technology, but the question remains, "What will be the market drivers?"

More on Friday...