is a speech-compression algorithm. It offers substantially better voice quality (even in noisy environment) because of doubled throughput, without extra radio and transmission bandwidth requirements.
It is standardized in 3GPP Rel-5 and applicable in 3GPP mobile circuit switched systems (GSM, WCDMA) as well as packet switched systems (IMS Telephony, VoIP).´
AMR-WB comprises nine coding rates including the first three rates 6.60, 8.85 and 12.65kbps, which make up the mandatory multirate configuration.
Most speech coding systems in use today are based on telephone-bandwidth narrowband speech, nominally limited to about 200-3400 Hz and sampled at a rate of 8 kHz. This limitation built into the public switched telephone network (PSTN) dates back to the first transcontinental telephone service at the beginning of the 20th century and imposes a constraint on communication quality.
Topics on this page
Technical Background
Comparison of Wideband Speech Coding Standards
Sample Sounds
Technical Highlights
Benefits
Applications
AMR-WB/G.722.2 Standard Specifications
File Format and RTP Packet Format Definitions
Today, the increasing penetration of end-to-end digital networks such as the second- and third-generation wireless systems (2G and 3G) and voice over packet networks permits the use of wider speech bandwidth.
This wider speech bandwidth offers communication quality significantly surpassing that of the PSTN in intelligibility and naturalness, giving an experience that is equivalent to face-to-face interaction.
Technical Background
The AMR-WB speech codec utilizes the ACELP® (Algebraic Code Excitation Linear Prediction) technology, which is also employed in the AMR narrowband and EFR speech codecs as well as in ITU-T G.729 and G.723.1 at 5.3 kbit/s, among others. The AMR-WB speech codec consists of nine speech codec modes with bit rates of 23.85, 23.05, 19.85, 18.25, 15.85, 14.25, 12.65, 8.85 and 6.6 kbps. AMR-WB also includes a background noise mode that is designed to be used in discontinuous transmission (DTX) operation in GSM and as a low bit rate source-dependent mode for coding background noise in other systems. In GSM the bit rate of this mode is 1.75 kbps.
Back to top
Comparison of Wideband Speech Coding Standards
Recommendation G.722 G.722.1 AMR-WB/G.722.2
Date 1988 1999 2001/2002
Bit rate (kbps) 48, 56, 64 (embedded) 24, 32 23.85, 23.05, 19.85 18.25, 15.85, 14.25, 12.65, 8.85, 6.6
Type Sub-band ADPCM Transform Coding Algebraic Code Excited Linear Prediction (ACELP®)
Delay (ms):
- Frame Size
- Lookahead
0.125
1.5
20
20
20
5
Quality Commentary
(at 64 kbps)
Poor speech performance in some operating conditions.
Scope of standard limited to hands-free and low packet loss rates.
Good music performance
Good speech performance at rates 12.65 kbit/s and higher:
15.85 >= G.722@56
23.05 >= G.722@64
Complexity 10 MIPS <15 MIPS 38 WMOPS
RAM (Kwords) 1 2 5.3
Fixed-point Bit exact Bit-exact C Bit-exact C
Floating-point None Exists in Annex B In ANSI-C code
VAD/DTX/CNG None None Included
Principal applications ISDN, video conferencing Same as G.722 plus VoPN Same as G.722.1 plus 3G wireless
Back to top
Sample sounds
Click here to hear a narrowband sample of “seed . . . feed . . . seed ”
Now click here to hear an AMR-WB audio sample of “seed . . . feed . . . seed”
(same sample filtered out above 2000 Hz)
Back to top
Technical Highlights
The AMR-WB codec includes:
A set of fixed-rate coding modes, which can be adapted according to network congestion, thus ensuring significant enhancement of QoS
With a bandwidth ranging from 50 to 7000 Hz, AMR-WB not only improves the intelligibility and naturalness of speech, but also adds a feeling of transparent communication and eases speaker recognition
Voice Activity Detector (VAD) functionality
Discontinuous Transmission (DTX) functionality in GSM and Source-Controlled Rate (SCR) functionality in 3G
In-band signaling for codec mode transmission
Bridge between wireline and wireless applications
Interoperability across different wireline and wireless applications
Back to top
Benefits
Wideband speech coding results in major subjective improvements in speech quality. Compared to narrowband telephone speech, low-frequency enhancement in AMR-WB from 50 to 200 Hz contributes to increased naturalness, presence, and comfort. The high-frequency extension from 3400 to 7000 Hz provides better fricative differentiation (for example, between words like fin and thin), and therefore higher intelligibility.
The adoption of AMR-WB by ETSI/3GPP and ITU-T (where it is referred to as G.722.2) is of significant importance because, for the first time, the same codec has been adopted for wireless as well as wireline services. This eliminates the need for transcoding and eases the implementation of wideband voice applications and services across a wide range of communication systems and platforms.
Not only does AMR-WB provide superior voice quality over the existing narrowband standards, but it is also very robust against transmission errors due to multi-rate operation and adaptation.
The AMR-WB package comprises not only the codec but also VAD, CNG, DTX (Discontinuous Transmission), payload and storage formats, and hooks to media file formats. AMR-WB provides end-to-end solutions to a variety of applications and is the only wideband technology offering such a complete range of built-in features.
Back to top
Applications
The naturalness of wideband speech coding is a significant feature in high-fidelity telephony and in extended telecommunications processes such as audio teleconferencing and program broadcasting. Application areas for wideband speech include:
Multimedia services for 3G mobile communication systems
Wideband telephony over packet networks
Audio and video teleconferencing
Internet applications such as:
– Broadcasting and streaming
– Chat and virtual reality immersion environments
– Multimedia real-time collaboration tools
– Archiving and distribution of narrative content
– Network-based language-learning applications
Digital radio broadcasting
VoIP
Compared to narrowband telephony, AMR-WB wideband speech coding delivers very noticeable improvements in speech quality and comprehensibility which are sure to please customers.
Digital Terminals
Digital handsets and digital terminals allow for improved sound quality because they are not tied to the traditional 4 kHz limitation. Equipment at both ends must share similar “hi-fi” capacity if they want to establish a call using a wideband payload. PCs, PDAs, cell phones, and IP phones are perfectly suited for the wideband experience.
Enterprise
An office is a perfect environment in which to implement a wideband experience, whether the environment is centralized in one building or distributed across remote offices linked by packet networks. A noteworthy feature of AMR-WB/G.722.2 is that it behaves like a narrowband codec when it connects to legacy systems to make calls to non-wideband devices.
Conferencing
Conferencing is one of the primary drivers for wideband speech communication. Conferencing emulates the face-to-face experience, and to achieve this goal, superior speech quality is required. Developers of conferencing equipment are among the early adoptors of wideband codecs.
Increasing On-net Traffic
VoIP system integrators have noticed that “interfacing” their VoIP cloud with the legacy PSTN requires a large amount of equipment (media gateways, SS7 gateways, echo cancellers). Wideband experience will boost the use of IP terminals on both ends, helping to reduce the number of:
Legacy terminals used by your customer base
“Outgoing” off-net calls to legacy terminals (thus increasing on-net IP-to-IP calls)
Back to top
AMR-WB/G.722.2 Standard Specifications
The AMR-WB codec was standardized by ETSI/3GPP in December 2000, and selected and approved by the ITU-T in July 2001 and January 2002, respectively. The ITU-T standard is referred to as G.722.2. The AMR-WB/G.722.2 wideband speech codec is defined, described, recommended and prescribed in these standards:
3GPP 2001 TS 26.171 gives a general overview of the AMR-WB standards, lists the related specifications and summarizes them.
http://www.3gpp.org/ftp/Specs/html-info/26171.htm
3GPP 2004 TS 26.111 recommends AMR-WB for (3G-324H) multimedia telephone handsets. http://www.3gpp.org/ftp/Specs/html-info/26111.htm
3GPP 2005 TS 26.140 requires MMS to support AMR-WB when wideband speech working at 16 kHz sampling frequency is supported.
http://www.3gpp.org/ftp/Specs/html-info/26140.htm
3GPP 2006 TS 26.141 requires IP Multimedia Subsystem (IMS) Messaging and Presence to support AMR-WB when wideband speech working at 16 kHz sampling frequency is supported.
http://www.3gpp.org/ftp/Specs/html-info/26141.htm
3GPP 2005 TS 26.234 requires packet-switched streaming (PSS) to support AMR-WB when wideband speech working at 16 kHz sampling frequency is supported.
http://www.3gpp.org/ftp/Specs/html-info/26234.htm
3GPP 2005 TS 26.235 requires packet-switched multimedia terminals at 16kHz and PoC terminals to support AMR-WB.
http://www.3gpp.org/ftp/Specs/html-info/26235.htm
3GPP (2005) TS 26.346 states that Multimedia Broadcast/Multicast Service (MBMS) clients that support wideband speech working at 16 kHz sampling frequency should/shall support the AMR-WB decoder.
http://www.3gpp.org/ftp/Specs/html-info/26346.htm
ITU-T 2002 G.722.2 recommends the codec as mandatory for wideband speech.
http://www.itu.int/rec/recommendation.asp?type=folders&lang=e&parent=T-REC-G.722.2
OMA (2005) Push to talk Over Cellular (PoC) “User Plane Specification” states the PoC server must support AMR-WB media parameters.
http://www.openmobilealliance.org/release_program/index.html
CableLabs® (2006) PKT-SP-CODEC-MEDIA-I01-060406, “PacketCable™ Codec and Media Specification,” includes AMR-WB as a supported wideband codec and notes that recommending the use of AMR-WB guarantees end-to-end wideband codec interoperability between User Equipment or Media Gateways and 3GPP cellular networks.
http://www.packetcable.com/downloads/specs/PKT-SP-CODEC-MEDIA-I01-060406.pdf
Back to top
File format and RTP packet format definitions
These specifications define the .3gp file format for use in packet-switched streaming services like MMS and PSS and the Real-Time Protocol (RTP) payload formats for packetizing AMR, AMR-WB, and AMR-WB+ encoded audio signals into the RTP. The RTP payload format definitions enable use of the codecs in RTP packet-switched networks in applications like streaming, and they enable interoperability with existing codec transport formats on non-IP networks.
3GPP (2005) TS 26.244, “Transparent end-to-end packet switched streaming service (PSS); 3GPP file format (3GP),” defines the 3GPP (.3gp) file format used by PSS and MMS and explains how AMR, AMR-WB and AMR-WB+ audio information can be encapsulated in .3gp files. http://www.3gpp.org/ftp/Specs/html-info/26244.htm
IETF (2002) RFC 3267, “Real-Time Transport Protocol (RTP) Payload Format and File Storage Format for the Adaptive Multi-Rate (AMR) and Adaptive Multi-Rate Wideband (AMR-WB) Audio Codec,” specifies the payload format for packetizing AMR and AMR-WB encoded speech signals into the RTP. The payload format is designed to enable using the codec in RTP packet-switched networks in applications like VoIP and streaming, and it enables interoperability with existing AMR and AMR-WB transport formats on non-IP networks. In addition, a file format is specified for transport of AMR and AMR-WB speech data in storage mode applications such as email. Two separate MIME type registrations are included, one for AMR and one for AMR-WB, specifying use of both the RTP payload format and the storage format. http://www.ietf.org/rfc/rfc3267.txt
Back to top