[FFmpeg-user] Building framemd5 checksum for Audio per second
Bauer, Christoph
Christoph.Bauer at slub-dresden.de
Wed Jul 24 14:43:52 EEST 2019
Dear Kieran O'Leary,
it worked! Thank you very much for your helpful addition.
All the best
Christoph
-----Ursprüngliche Nachricht-----
Von: ffmpeg-user <ffmpeg-user-bounces at ffmpeg.org> Im Auftrag von Kieran O Leary
Gesendet: Mittwoch, 24. Juli 2019 11:12
An: FFmpeg user questions <ffmpeg-user at ffmpeg.org>
Betreff: Re: [FFmpeg-user] Building framemd5 checksum for Audio per second
Hi,
On Wed, Jul 24, 2019 at 9:46 AM Bauer, Christoph < Christoph.Bauer at slub-dresden.de> wrote:
> Hello everybody,
>
> I was building framemd5-checksums for Audio with 96kHz sampling for
> every second, using this command line:
>
> ffmpeg -i INPUT.mka -filter_complex "asetnsamples=n=96000" -f framemd5
> OUTPUT.framemd5
>
> The framemd5-file I get out of this starts with these lines:
>
> #format: frame checksums
> #version: 2
> #hash: MD5
> #software: Lavf58.27.102
> #tb 0: 1/96000
> #media_type 0: audio
> #codec_id 0: pcm_s16le
> #sample_rate 0: 96000
> #channel_layout 0: 3
> #channel_layout_name 0: stereo
> #stream#, dts, pts, duration, size, hash
> 0, 0, 0, 96000, 384000,
> 1fd71ae8708c0b45e229b1d823eb362b
> 0, 96000, 96000, 96000, 384000,
> 2b9a626285c754e3b39d1555f202ad81
> 0, 192000, 192000, 96000, 384000,
> 8e9a791e7d5126e5208416f412f3ab26
>
> The same file was delivered with a framemd5-file which starts with
> those
> lines:
>
> #format: frame checksums
> #version: 2
> #hash: MD5
> #software: Lavf58.20.100
> #tb 0: 1/96000
> #media_type 0: audio
> #codec_id 0: pcm_s24le
> #sample_rate 0: 96000
> #channel_layout 0: 3
> #channel_layout_name 0: stereo
> #stream#, dts, pts, duration, size, hash
> 0, 0, 0, 96000, 576000,
> 2b31df8f73cf3d95b236d2dd16e6ba30
> 0, 96000, 96000, 96000, 576000,
> ebb8f1f3efde3fda9cb21ecd959ea076
> 0, 192000, 192000, 96000, 576000,
> 1a01031d660895b43f92234f2582a6a8
>
> The difference between the checksum-files is as far that i can see at
> first the software (Lavf58.20.100 vs. Lavf58.27.102), at second the
> codec_id (pcm_s24le vs. pcm_s16le) and thirdly the size (576000 vs.
> 384000). Now my questions are the following:
>
> The version of lavf most likely isn't the issue, but the audio format
would be why there's different framemd5s and sizes.
>
> - Is it possible to get out those different framemd5-files out
> of the same file?
>
Probably - ffmpeg defaults to decoding as 16-bit for framemd5 for audio. So whoever made the 24-bit framemd5 probably added `-c:a pcm_s24le` to their
framemd5 command. You should add the same to your framemd5 command.
I don't know if this will produce the same framemd5s or not, but I've had to force pix_fmt sometimes in order to get matching image framemd5s.
>
> - Which parameters have to be added or changed in the command
> line above to get the result oft he second framemd5-file?
>
by adding `-c:a pcm_s24le` this time?
ffmpeg -i INPUT.mka -filter_complex "asetnsamples=n=96000" -c:a pcm_s24le -f framemd5 OUTPUT.framemd5
This is just what I'd try, curious to know if it would work.
Best,
Kieran O'Leary
Irish Film Institute.
>
>
> Thanks in advance
>
> Christoph
>
>
> _______________________________________________
> ffmpeg-user mailing list
> ffmpeg-user at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-request at ffmpeg.org with subject "unsubscribe".
_______________________________________________
ffmpeg-user mailing list
ffmpeg-user at ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email ffmpeg-user-request at ffmpeg.org with subject "unsubscribe".
More information about the ffmpeg-user
mailing list