ITU-T Standard Vocoders
G.729/G.729B
The algorithms implemented are
the ITU-T G.729 and G.729B recommendations, fixed rate speech coders for internet
and multimedia communications. G.729 compresses narrowband speech and audio
signals at a sample rate of 8kHz to 8 000 bps,
using a linear predictive analysis-by-synthesis coding, Conjugate-Structure
Alegraic-Code-Excited Linear Prediction (CS-ACELP). The coder operates on
10ms frames to extract the parameters of the CELP coding model. The decoder
uses two filters. One is based on a 10th order Linear Prediction (LP) filter,
the other is based on an adaptive-codebook approach. Additional performance
enhancement is achieved by use of a Post Filter.
G.729B includes G.729 with the
additional option of Annex B. Annex B implements silence compression techniques
to reduce the transmitted bit rate during the silent intervals of speech and
conforms to Recommendation V.70 silence compression. Systems allowing discontinuous
transmission (DTX) are based on Voice Activity Detection (VAD) algorithms
and Comfort Noise during the silence intervals. This also provides the additional
advantage of usig lower processing loads and DSP bandwidth resource during
silence frames.
back to menu