Hi,
I'm looking for a way to process audio from a GMR-1 (Thuraya) network. I know the codec is done by dvsinc and based on AMBE, at a 4kbps bitrate.
From the specs :
===
The speech encoder takes 16-bit uniform pulse code modulation (PCM) samples as input. One frame of encoded speech consists of two quantizer-frames, which each contain 80 bits. The encoded speech at the output of the voice encoder is delivered to the channel coding function defined in GMR-1 05.003 [2] to produce an encoded frame consisting of two quantizer-frames, each containing 104 bits. The vocoder therefore produces a gross bit rate of 5,2 kbps where 4,0 kbps are used for voice data and the remaining 1,2 kbps are used for error control.
===
The error control is known. So what's I'm looking for is how to convert the 80 bits (48 first class bits per quantizer-frame and 32 second class bits per quantizer-frame) into audio and vice & versa.
Any help is welcome.
Cheers,
Sylvain
I'm looking for a way to process audio from a GMR-1 (Thuraya) network. I know the codec is done by dvsinc and based on AMBE, at a 4kbps bitrate.
From the specs :
===
The speech encoder takes 16-bit uniform pulse code modulation (PCM) samples as input. One frame of encoded speech consists of two quantizer-frames, which each contain 80 bits. The encoded speech at the output of the voice encoder is delivered to the channel coding function defined in GMR-1 05.003 [2] to produce an encoded frame consisting of two quantizer-frames, each containing 104 bits. The vocoder therefore produces a gross bit rate of 5,2 kbps where 4,0 kbps are used for voice data and the remaining 1,2 kbps are used for error control.
===
The error control is known. So what's I'm looking for is how to convert the 80 bits (48 first class bits per quantizer-frame and 32 second class bits per quantizer-frame) into audio and vice & versa.
Any help is welcome.
Cheers,
Sylvain
