G.722.2 | GSM-AMR Wideband Codec

The Adaptive Digital Technologies G.722.2 Audio Coder is a real-time implementation of the ITU G.722.2 audio coder also referred to as the Adaptive Multi-Rate Wideband (AMR-WB) codec. Wideband speech coding is a substantial feature in high-fidelity telephony and in telecommunications processes such as audio teleconferencing and program broadcasting. The G.722.2 audio coder encodes 16 kHz sampled audio signals for transmission, and compresses the signals to a multitude of bit rates ranging from 6.6 kbit/s to 23.85 kbit/s. The G.722.2 coder provides 7 kHz of audio bandwidth.

AMR-WB provides improved speech quality due to a wider speech bandwidth of 50–7000 Hz compared to narrowband speech coders which in general are optimized for traditional POTS line quality of 300–3400 Hz.

G.722.2 Adaptive Multi-Rate Wideband is the first wideband codec to be standardized for both wireless (3GPP) and wireline (ITU-T Recommendation G.722.2) applications.

OVERVIEW

The ITU-T Recommendation G.722.2, Wideband coding of speech at around 16 kbit/s using Adaptive Multi-Rate Wideband (AMR-WB), was approved in July of 2003 by the ITU-T Study Group 16 under the ITU-T Recommendation A.8 procedure. This proved to be a significant event throughout international telecommunications communities, because the same specification was also adopted by the 3 rd Generation Partnership Project (3GPP). The standardization of AMR-WB by both ITU and 3GPP meant that the same algorithms and processing techniques can be deployed on traditional LAN line circuits, as well as wireless communication/radio networks.

The AMR Wide Band codec is based on the Code Excited Linear Prediction (CELP) coding model, allowing the end user network to experience truer voice signals due to the higher bandwidth. Compared to typical narrow band coding, the AMR-WB codec operates at double the sample rate which ultimately leads to more natural sounding voice and music content. The codec also employs other routines that have become characteristic to telephony networks such as Voice Activity Detection (VAD), Discontinuous Transmission (DTX) and Comfort Noise Generation (CNG).