Telephony networks employ a system of coders and decoders, called a codec. A codec is used to reduce bandwidth requirements over limited capacity channels in two-way real-time communications. The codec takes an analog voice signal from an input source, like a microphone, and converts the analog signal into a digital format that can be transmitted across a packet network. At the receiving end the signal is converted back again to an analog signal for human consumption.
The G.7xx family of standards are comprised of speech and audio codecs that are primarily used in cellular telephony and Internet telephony including VoIP communications. There are several specifications (protocols) in the G.7xx family of G Series ITU Recommendations, including the following:
G.711 – Also known as Pulse Code Modulation (PCM), it is the ITU-T international standard for encoding telephone audio on a 64 kbps channel. PCM samples the signal 8000 times a second; each sample is represented by 8 bits for a total of 64 kbit/s. There are two versions of the this standard codec. The -law (pronounced as mew law) is generally used in North America and Japan digital communications. The A-law is used in European digital communications. The difference between the two standards is the method in which the analog signal is sampled. (See also PCM).
G.721 – An ITU-T standard codec that uses Adaptive Differential Pulse Code Modulation (ADPCM); a form of Pulse Code Modulation (PCM), to produce a digital signal with a lower bit rate than standard PCM. This ITU standard for speech codecs uses ADPCM on a 32 kbit/s channel. NOTE: G.721 was first introduced in 1984. In 1990 this standard was folded into G.726 along with G.723. (See also ADPCM).
G.722 – An ITU-T standard codec that uses sub-band adaptive differential pulse code modulation (SB-ADPCM) within a bit rate of 64 kbit/s. The system is referred to as 64 Kbps (7 kHz) audio coding. SB-ADPCM splits the frequency band into two sub-bands (higher and lower) and the signals in each sub-band are encoded using ADPCM. Extensions to the G.722 standard include the following;
- G.722.1 – Is the ITU-T standard for low-complexity coding at 24 and 32 kbit/s for hands-free operation in systems with low frame loss.
- G.722.2 – Is the ITU-T standard for coding at 24 and 32 kbit/s for hands-free operation in systems with low frame loss.
G.723 – An ITU-T standard codec that uses Adaptive Differential Pulse Code Modulation (ADPCM) standard for speech codecs on a 24 and 40 kbps channel. NOTE: G.723 was first introduced in 1988. In 1990 this standard was folded into G.726 along with G.721. (See also ADPCM).
G.726 – An ITU-T Adaptive Differential Pulse Code Modulation (ADPCM) standard speech codec used for the transmission of voice at rates on 16, 24, 32, and 40 kbit/s channels. G.726 supersedes both G.721 and G. 723 as it includes both of these standards plus includes the new standard for the 16 kbit/s rate. G.726 was the standard codec used in Digital Enhanced Cordless Telecommunications (DECT) wireless phone systems. (See also ADPCM).
G.727 – A specialized version of the ITU-T G.726 protocol that is intended for packet-based systems using the Packetized Voice Protocol (PVP). G.727 uses 5, 4, 3 and 2-bit/sample embedded adaptive Differential Pulse Code Modulation (ADPCM). (See also ADPCM).
G.728 – An ITU-T speech coding standard that uses Low Delay Code Excited Linear Prediction (LD-CELP) operating at 16 kbit/s compression at a sampling rate of 8,000 samples per second. The algorithmic coding delay of G.728 is 0.625 ms. G.728, when compared to G.726 delivers close to the same voice quality but uses only one-half the bandwidth.
G.729 – An ITU-T audio data compression standard that operates at 8 kbit/s using a conjugate-structure algebraic-code-excited linear-prediction (CS-ACELP). This algorithm for voice compresses voice audio in 10 millisecond frames. G.729 is commonly used in in Voice over IP (VoIP) applications because of its inherently low bandwidth requirement. Extensions to the G.729 standard include the following;
- G.729a (G.729 Annex A) – Compatible with G.729 Annex A specifies a coder with several simplifications, including code book search routines. These modifications are known to often result in a slightly lower voice quality.
- G.729b (G.729 Annex B) – Compatible with G.729, Annex B specifies a coder that uses Discontinuous Transmission (DTX), Voice Activity Detection (VAD), and Comfort Noise Generation (CNG) to reduce bandwidth usage. Bandwidth is reduced by preventing the transmission of non-voice during periods of silence.