adaptive multirate (compression) (AMR)

Adaptive Multirate (AMR) is a variable bit ratevoice codec standardized by the Third Generation Partnership Project(3GPP) and the European Telecommunications Standards Institute( ETSI), which is used in the GSM standard and in 3rd generation( 3G) mobile networks for voice compression.

AMR compression is characterized by a high compression rate. For compression, AMR uses various compression algorithms such as Algebraic Code Excited Linear Prediction( ACELP), Discontinuous Transmission( DTX), Voice Activity Detection( VAD) and Comfort Noise Generation( CNG).

There are two AMR versions: AMR- NB, which stands for Narrow Band, and AMR- WB for Wideband. Both versions are compatible with each other. The narrowband version AMR-NB is mainly used in the GSM standard, while AMR-WB is used in UMTS.

AMR-NB is for the speech range from 300 Hz to 3.4 kHz and has as speech codec a combination of several speech codecs, which works with a sampling rate of 8 kHz and a sampling depth of 16 bits. The data rate adjusts in eight steps between 4.75 kbit/s and 12.2 kbit/s to the transmission quality of the connection. At the highest data rate, the AMR codec also has the highest speech quality. It corresponds to that of the EFR method, Enhanced Full Rate (EFR). Lower data rates have lower speech quality, but more bits can be invested in error correction, which means that intelligible communication can still take place even with a high bit error frequency.

AMR-WB was developed by Nokia and VoiceAge and specified by 3GPP. Later, the International Telecommunication Union( ITU) standardized wideband voice compression under G.722. AMR-WB covers the voice frequency range between 50 Hz and 7 kHz, which makes speech sound much more natural. There are several modes with bit rates between 6.6 kbit/s and 23.85 kbit/s. Bit rates up to 19.85 kbit/s are also supported by the GSM standard. AMR-WB technology is used, among other things, in mobile networks to transmit HD voice, which results in a significant improvement in voice quality. During transmission, the network dynamically selects the optimal mode and data rate; a low data rate is selected when traffic is heavy, and a high data rate is selected when traffic is light.

The upcoming technology to improve voice quality is Enhanced Voice Service( EVS), which operates at a bandwidth of 20 kHz.