FreeCalypso > hg > gsm-codec-lib
annotate doc/EFR-testing @ 605:63f774192906
gsmhr_decoder_twts002_in(): set BFI=1 SID=1 for invalid SID
When a received TW-TS-002 RTP payload indicates invalid SID,
which of the 3 possible BFI/SID combinations should we pass to
our internal ETSI-based speech decoder or TFO engine?
Our original code passed BFI=0 SID=1, but upon further reflection,
BFI=1 SID=1 is a better choice. In the corner case where received
invalid SID is fed to a full decoder in homed state, setting BFI=1
allows that decoder to emit zeros on PCM and stay homed, instead of
launching into full decoding.
| author | Mychaela Falconia <falcon@freecalypso.org> |
|---|---|
| date | Thu, 04 Dec 2025 21:01:46 +0000 |
| parents | 1e8569000049 |
| children |
| rev | line source |
|---|---|
|
132
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
1 When it comes to codec libraries, testing for correctness is essential, and EFR |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
2 is no exception. There is a set of EFR encoder and decoder test sequences |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
3 published by ETSI in ts_100725v050200p0.zip (GSM 06.54), and our suite of tools |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
4 includes gsmefr-etsi-enc and gsmefr-etsi-dec test programs that operate on the |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
5 representation formats used by these test sequences. Because these test |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
6 programs are based on libgsmefr EFR_encode_frame() and EFR_decode_frame() |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
7 functions, seeing gsmefr-etsi-enc produce output that matches official ETSI |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
8 *.cod files proves that libgsmefr encoder is correct, and seeing gsmefr-etsi-dec |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
9 produce output that matches official ETSI *.out files proves that libgsmefr |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
10 decoder is correct. |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
11 |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
12 For debugging, we also have gsmefr-cod-parse and gsmefr-dec-parse utilities that |
|
3fd48946dcb8
doc/EFR-testing article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
13 parse ETSI *.cod and *.dec file formats and dump their content in human-readable |
|
134
170e03b20337
doc/EFR-testing (gsmefr-{cod,dec}-parse): document similarity to gsmrec-dump
Mychaela Falconia <falcon@freecalypso.org>
parents:
132
diff
changeset
|
14 form similar to gsmrec-dump. |
|
158
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
15 |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
16 Please note that all ETSI test sequence file formats are endian-dependent: their |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
17 original programs read and write 16-bit words in the local machine's native byte |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
18 order, and whenever you are working with published test sequence files, you have |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
19 to check to see if they are BE or LE. Our gsmefr-etsi-{enc,dec} and |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
20 gsmefr-{cod,dec}-parse programs support both byte orders; the default is LE |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
21 (matching the main parts of ts_100725v050200p0.zip), or you can select BE with |
|
1e8569000049
doc/EFR-testing: document endian issues
Mychaela Falconia <falcon@freecalypso.org>
parents:
134
diff
changeset
|
22 -b option. |
