[MPlayer-users] DVB stream including LATM AAC Audio issues

[Steven Ellis]

> Il Sunday 06 January 2008 23:48:45 hai scritto:
>> Il Saturday 05 January 2008 00:14:53 Nico Sabbi ha scritto:
>> > Il Thursday 03 January 2008 21:30:49 Nico Sabbi ha scritto:
>> > > Il Thursday 03 January 2008 21:21:18 Steven Ellis ha scritto:
>> > > > Ok I have a link to an alternative wrapper that provides the
>> > > > same functionality where they have published the source code.
>> > > >
>> > > > http://blog.monogram.sk/janos/2008/01/03/monogram-aac-decoder
>> > > >-v 09 00 /
>> > >
>> > > thanks, I'll see to adapt it to mplayer during the weekend
>> >
>> > I can finally see that the streams are conforming to my parser.
>> > The major problem is that I'll probably need to patch
>> > the internal faad because otherwise I can't parse the
>> > whole configurator. Stay tuned
>>
>> talking about LATM:
>>
>> here's a very first & dirty version, but be warned that:
>> a) it's supposed to work on a very limited subset
>> of the possible formats: the one supposed to be used
>> in DVB
>> b) if it works (not for me: too little data to work on)
>> it's almost a miracle :)
>> c) it's not even remotely in the right format: the
>> best choice is patching faad
>> d) use only the internal faad, or compilation will break
>>
>> If you test it on some (possibly long) sample you will do
>> me a favor; patches welcome :)
>
> forgotten patch attached

Thanks for the patch. I've tried it on my 1 minute sample videos from NZ
with limited luck

tv6.ts produces no audio

tv7.ts produces a couple of seconds of audio and then crashes with the
following backtrace
Error while decoding frame!
mplayer: mdct.c:63: faad_mdct_init: Assertion `N % 8 == 0' failed.


MPlayer interrupted by signal 6 in module: decode_audio
- MPlayer crashed. This shouldn't happen.
  It can be a bug in the MPlayer code _or_ in your drivers _or_ in your
  gcc version. If you think it's MPlayer's fault, please read
  DOCS/HTML/en/bugreports.html and follow the instructions there. We can't
and
  won't help unless you provide this information when reporting a possible
bug.
*** glibc detected *** ./mplayer: free(): invalid pointer: 0x00120012 ***
======= Backtrace: =========
/lib/libc.so.6[0x8a88b6]
/lib/libc.so.6(cfree+0x90)[0x8abe00]
./mplayer(NeAACDecClose+0x33)[0x84cfb53]
./mplayer(uninit_audio+0xd4)[0x80df2a4]
./mplayer(uninit_player+0x5e)[0x8078e4e]
./mplayer(exit_player_with_rc+0x2e)[0x807943e]
./mplayer(exit_player+0x1a)[0x807963a]
[0xac9420]
/lib/libc.so.6(abort+0x101)[0x86c521]
/lib/libc.so.6(__assert_fail+0xfb)[0x86425b]
./mplayer(faad_imdct+0x0)[0x84d1ae0]
./mplayer(filter_bank_init+0x3c)[0x84d057c]
./mplayer(NeAACDecInitFromBitfile+0x109)[0x84cfcf9]
./mplayer[0x811beed]
./mplayer(decode_audio+0x10d)[0x80def2d]
./mplayer(main+0x1852)[0x807b802]
/lib/libc.so.6(__libc_start_main+0xdc)[0x857dec]
./mplayer(vfprintf+0x421)[0x80775c1]


sport.ts also crashes out but with rather less verbose output

mplayer: mdct.c:63: faad_mdct_init: Assertion `N % 8 == 0' failed.


MPlayer interrupted by signal 6 in module: decode_audio
- MPlayer crashed. This shouldn't happen.
  It can be a bug in the MPlayer code _or_ in your drivers _or_ in your
  gcc version. If you think it's MPlayer's fault, please read
  DOCS/HTML/en/bugreports.html and follow the instructions there. We can't
and
  won't help unless you provide this information when reporting a possible
bug.


MPlayer interrupted by signal 11 in module: uninit_acodec
Segmentation fault


I can't try the patch out on a live stream at the moment but I might have
a go tonight.

Steve