[FFmpeg-user] ffmpeg wav lossless to youtube mov?

[John Hawkinson]

Zathras: It appears your assumption (6) that the files are "exactly the
same" because the spectrograms are the same is not correct.

I downloaded https://www2.cs.uic.edu/~i101/SoundFiles/BabyElephantWalk60.wav
as a test file and converted it with the freeoconvert.com online converter,
and ffprobe reports the following:

% ffprobe BabyElephantWalk60.mov
...
  Stream #0:1[0x2]: Audio: aac (LC) (mp4a / 0x6134706D), 22050 Hz, mono,
fltp, 72 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
      vendor_id       : [0][0][0][0]

That is, the online converter you specify has converted the audio to AAC.
That is, you should be able to use simply

ffmpeg -i picture.jpg -i music.wav upload.mov

and accept the results.

Alternatively, discussion in this thread suggests you can use the FLAC
encoding, which is lossless.

I agree that the responses you received to your inquiry were rude and
unprofessional and have no place on this mailing list and I am sorry you
had to deal with them, and I do not understand who administers this list or
how to resolve those issues.
--
jhawk at alum.mit.edu
John Hawkinson


On Thu, May 23, 2024 at 7:12 PM Zathras Yes <zathras.the.first at gmail.com>
wrote:

> Hi,
>
> as i have mentioned to Mr. Reindl, the online-converter
> https://www.freeconvert.com/wav-to-mov  does the trick.
>
> ffmpeg should be able to do the same, I suppose.
> May be, I am in error, may be not...
>
> 1. make a wav-files (16 Bit, 44.1 kHz) "sound.wav"
> 2. use the mentioned online converter from "sound.wav" to "upload.mov"
> 3. Upload this "upload.mov" to youtube.
> 4. Go to the youtube-link and download the file "download.mov"
> 5. extract the sound file out of "download.mov" --> "extracted.wav"
> 6. Compare the original file "sound.wav" with "extracted.wav" with its
> spectrogramms. (Audacity or Spectral Layers for instance)
> --> et voila the files are exactly the same!
>
> So, it does work somehow.
> For this, I do not understand the length of the discussion (especially with
> Mr. Reindl), that it does not work, because it is not supposed to work?
> But it does -  so I wonder why it is such a big deal here.
>