[FFmpeg-user] Different audio rates on input DV files

[Carl Eugen Hoyos]

2019-01-16 21:10 GMT+01:00, Bill Moseley <moseley at hank.org>:
> On Tue, Jan 15, 2019 at 6:27 PM Bill Moseley <moseley at hank.org> wrote:
>
>
>> $ ffmpeg -f concat -safe 0 -i input.txt -vf yadif out.mp4
>>
>> Apparently audio settings on the camera changed during that day.
>>
>> $ ffmpeg -i 'clip-1999-07-29 13;25;19.dv' 2>&1 | grep Audio
>>     Stream #0:1: Audio: pcm_s16le, 32000 Hz, stereo, s16, 1024 kb/s
>>     Stream #0:2: Audio: pcm_s16le, 32000 Hz, stereo, s16, 1024 kb/s
>>
>> $ ffmpeg -i 'clip-1999-07-29 13;26;20.dv' 2>&1 | grep Audio
>>     Stream #0:1: Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s
>>
>>
> I guess the answer is in the documentation:
> https://ffmpeg.org/ffmpeg-formats.html#concat
>
> "All files must have the same streams (same codecs, same time base, etc.)."
>
>
> Looks like I need to normalized the dv files first, so any that are not
> 48000 Hz:
>
> $ ffmpeg -i input.dv -ar 48000 out.dv

$ ffmpeg -i input.dv -ar 48k -vcodec copy out.dv

The alternative is to use the concat filter that does not
have the format limitations.

Carl Eugen