Hi,
This issue should be fixed in git master and in 1.1-beta, which will
hopefully be released very soon.
Cheers,
Jean-Marc
On 11/07/13 04:57 PM, Marc Lindahl wrote:> I didn't expect this, is it normal? When you use inbandfec, and have
packet losses, you end up with more audio samples than you started with.
>
> With short frame (2.5), the FEC isn't supposed to do anything but it
shouldn't do that should it? and 20mS frame, should do 'something'
but? add audio?
>
> I'm using opus_demo for this.
>
> Here's the steps (I have my test audio in it's own folder, parallel
with opus and sox):
>
> make raw audio:
>
> ../sox-14.4.1/sox music_orig.wav -t raw music_orig.raw
>
> try a few different encodings:
>
> ../opus-1.1-alpha/opus_demo audio 48000 2 128000 -framesize 2.5 -loss 10
-inbandfec music_orig.raw music_orig_f2.5_b128000_lr10_fec.raw
> ../opus-1.1-alpha/opus_demo audio 48000 2 128000 -framesize 2.5 -loss 10
music_orig.raw music_orig_f2.5_b128000_lr10.raw
> ../opus-1.1-alpha/opus_demo audio 48000 2 128000 -framesize 20 -loss 10
-inbandfec music_orig.raw music_orig_f20_b128000_lr10_fec.raw
> ../opus-1.1-alpha/opus_demo audio 48000 2 128000 -framesize 20 -loss 10
music_orig.raw music_orig_f20_b128000_lr10.raw
>
> (note - sometimes you get 'error decoding frame: buffer too small'
message?)
>
> check file sizes:
>
> ls -lsh music_orig*
> 34056 -rw-r--r-- 1 mlindahl staff 17M Jul 11 15:53 music_orig.raw
> 34056 -rw-r--r-- 1 mlindahl staff 17M Sep 4 2012 music_orig.wav
> 34048 -rw-r--r-- 1 mlindahl staff 17M Jul 11 15:59
music_orig_f2.5_b128000_lr10.raw
> 179408 -rw-r--r-- 1 mlindahl staff 88M Jul 11 15:57
music_orig_f2.5_b128000_lr10_fec.raw
> 34048 -rw-r--r-- 1 mlindahl staff 17M Jul 11 16:49
music_orig_f20_b128000_lr10.raw
> 49192 -rw-r--r-- 1 mlindahl staff 24M Jul 11 16:01
music_orig_f20_b128000_lr10_fec.raw
>
>
> Notice the FEC files are significantly longer. They also sound very odd,
like frames are being repeated?
>
> Im I doing something wrong or?.?
>
> Regards,
> Marc
> _______________________________________________
> opus mailing list
> opus at xiph.org
> http://lists.xiph.org/mailman/listinfo/opus
>