opus - Jun 2022 - Antw: [EXT] Glitch in the ope_encoder_drain();

>>> m b <mvetnk at email.com> schrieb am 09.06.2022 um 11:02 in
Nachricht
<trinity-ac212bc6-5632-4a70-98a5-65076eac5566-1654765371074 at
3c-app-mailcom-lxa02
:
> I found there is some text information at the end of the wave file, that 
> caused the cracking sound in the opus file, now after proper processing the
> raw data, the problem is sovled. Thank you Ulrich for your help!
Glad that my initial suspect was correct ;-)
> Sent: Tuesday, June 07, 2022 at 6:30 PMFrom: "Ulrich Windl" 
> <Ulrich.Windl at rz.uni-regensburg.de>To: mvetnk at email.comSubject:
Re: Antw:
> [EXT] [opus] Glitch in the ope_encoder_drain();
>>>> m b <mvetnk at email.com> schrieb am 07.06.2022 um 11:25
in
> Nachricht<trinity-c4b70144-339f-4410-9570-6d06a273a3c9-1654593936399 at
3c-app-m
> ailcom-lxa10:> ALL the end part of the WAVE is audio raw data, and the
end of
> the decoded> Opus is cracking sound. Thank you!Well, I expected the WAV
to be
> RAW data; what I wanted to know or see is whether the last sample is at
zero.>
> Sent: Tuesday, June 07, 2022 at 2:37 PMFrom: "Ulrich Windl"> 
> <Ulrich.Windl at rz.uni-regensburg.de>To: mvetnk at email.com, opus
at xiph.orgSubject:>
> Antw: [EXT] [opus] Glitch in the ope_encoder_drain();> How does the end
of
> your WAV look like, and how does the end of the decoded> Opus look
like?(Maybe
> you can show waveform displays)

On Thu, Jun 9, 2022 at 5:23 AM Ulrich Windl <
Ulrich.Windl at rz.uni-regensburg.de> wrote:
> >>> m b <mvetnk at email.com> schrieb am 09.06.2022 um 11:02
in Nachricht
>
> <trinity-ac212bc6-5632-4a70-98a5-65076eac5566-1654765371074 at
3c-app-mailcom-lxa02
> :
> > I found there is some text information at the end of the wave file,
that
> > caused the cracking sound in the opus file, now after proper
processing
> the
> > raw data, the problem is sovled. Thank you Ulrich for your help!
>
> Glad that my initial suspect was correct ;-)
>
We had the same bug in speexenc some time ago, for WAV files with comments
at the end you'd hear an audible glitch.

Here's the relevant fix in case it's useful to anyone:
https://gitlab.xiph.org/xiph/speex/-/commit/566a73a6c1e28c13979f26732e55c4f3629689e8


Best,
-t

>
> > Sent: Tuesday, June 07, 2022 at 6:30 PMFrom: "Ulrich Windl"
> > <Ulrich.Windl at rz.uni-regensburg.de>To: mvetnk at
email.comSubject: Re:
> Antw:
> > [EXT] [opus] Glitch in the ope_encoder_drain();
> >>>> m b <mvetnk at email.com> schrieb am 07.06.2022 um
11:25 in
> >
> Nachricht<trinity-c4b70144-339f-4410-9570-6d06a273a3c9-1654593936399 at
3c-app-m
> > ailcom-lxa10:> ALL the end part of the WAVE is audio raw data, and
the
> end of
> > the decoded> Opus is cracking sound. Thank you!Well, I expected the
WAV
> to be
> > RAW data; what I wanted to know or see is whether the last sample is
at
> zero.>
> > Sent: Tuesday, June 07, 2022 at 2:37 PMFrom: "Ulrich
Windl">
> > <Ulrich.Windl at rz.uni-regensburg.de>To: mvetnk at email.com,
> opus at xiph.orgSubject:>
> > Antw: [EXT] [opus] Glitch in the ope_encoder_drain();> How does the
end
> of
> > your WAV look like, and how does the end of the decoded> Opus look
> like?(Maybe
> > you can show waveform displays)
>
>
>
> _______________________________________________
> opus mailing list
> opus at xiph.org
> http://lists.xiph.org/mailman/listinfo/opus
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.xiph.org/pipermail/opus/attachments/20220609/f31310c7/attachment.htm>