EVRC-B

Speech Codec B

Enhanced Variable Rate Coder  

board

Features List

  • Functions are C-callable.
  • Multi–Channel capable.
  • Can be integrated with echo cancellers, and tone detection/regeneration.
  • Encodes frames of 20 ms sampled at 8KHz (160 samples) using above algorithms.
  • Bit-exact with all EVRCB 3GPP2 test sequences
  • The implementation support run time table relocation
  • Fully interruptible Code
  • Efficient scratch memory management with reduced stack requirements
  • The C55x implementation is xDAIS/xDM1.0 complaint

Coding Rate: 8.5, 4.0, 2.0, and 0. 8 kbits 
Sampling Rate: 8 kHz
Delay: 20 milliseconds

Availability

Platforms
Texas Instruments – TI TMS320C6000 C64x / C64x+ / C66x || TMS320C5000 C55x
Win32 Static Lib / DLL

ADT EVRC is available on the above Platforms: Other configurations are available upon request.

Specifications

NOTE: We specify MIPS (Millions of Instructions Per Second) as MCPS (Millions of Instruction Cycles Per Second). Unless otherwise specified, peak MIPS are indicated.

EVRC-B C64x

CPU Utilization & Memory Requirements
All Memory usage is given in units of byte.
SoftwareMIPS – Peak LoadingProgram MemoryData MemoryScratchPer Channel Data Memory
EVRC-B281201.8KB7.46KB30042176

EVRC-B C55x

CPU Utilization & Memory Requirements
All Memory usage is given in units of byte.
SoftwareMEGACYCLES /SEC*Program MemoryData Memory
Encode23.183.9445.81
Decode8.153.0944.46
Encode/Decode31.2118.748.19

*Measured with all program, constant tables, instance and CDP in SARAM, scratch in DARAM. All performance statistics with DTMF and TTY flags turned off.

EVRC-B Win32

CPU Utilization & Memory Requirements
All Memory usage is given in units of byte.
SoftwareMIPS – Peak LoadingProgram MemoryData MemoryScratchPer Channel Data Memory
Encode22922807244
Decode5730042180
Encode/Decode306357KB41KB

Description

Enhanced Variable Rate Coder B (EVRC-B) is a speech codec used by CDMA (Code Division Multiple Access) networks. EVRC-B is an extension to EVRC and compresses each 20 millisecond FRAMES of 8000 Hz, 16-bit sampled speech input into output frames of one of the four different sizes: Rate 1 – 171 bits/frame, Rate 1/2 – 80 bits/frame, Rate 1/4 – 40 bits/frame, Rate 1/8 – 16 bits/frame.

One significant enhancement in EVRC-B vocoder is the use of 1/4-rate frames that were not used in EVRC. The EVRC-B makes use of the intermediate coding rates through increased awareness of the nature of the individual speech samples. This provides lower average data rates compared to EVRC, for a given voice quality.

The EVRC-B decoder works on various rate packets, performs FEC and generates 20msec speech data frames. With a sampling frequency of 8000Hz the decoder output frame contains 160 input samples. The format of the output samples is assumed to be 16 bit uniformly quantized (linear).  

Function APIs

API function call summary

InitializeEVRC_BEncoderChannel (. . .)      Initializes the EVRC-B encoder 
InitializeEVRC_BDecoderChannel (. . .)      Initializes the EVRC-B decoder
EncodeEVRC_B (. . .)      Executes the EVRC-B encoder
DecodeEVRC_B (. . .)      Executes the EVRC-B decoder

Applications

Wireless/Mobile/Personal Communications (base-stations and handset)

 

Translate »