speech codec

Harald Welte laforge at gnumonks.org
Mon Jul 25 07:39:55 UTC 2011

On Sun, Jul 24, 2011 at 01:02:40PM +0100, Graham Palmer wrote:

> I too was wondering how the u-plane data was passed to the codec.
> 300-395-2 states that acelp-codec requires 2 * (137+1) 16bit samples per frame
> and not 216 + 216 bits indicated in the tch reordering file comments.

This is the input to the codec, i.e the PCM audio data.

the 216 + 216 bits are the encoded data, i.e. the codec output (which is
the decoder input).

The re-ordering code in osmo-tetra is untested.  You can feed the raw
burst bits into the ETSI reference codec, as it includes the error
correction / viterbi decoder.

- Harald Welte <laforge at gnumonks.org>           http://laforge.gnumonks.org/
"Privacy in residential applications is a desirable marketing option."
                                                  (ETSI EN 300 175-7 Ch. A6)

More information about the tetra mailing list