On 10/04/2018 11:40 AM, Nicolas George wrote:
Hi.
No need to Cc people who are on the list.
Actually replied only to you initially, sorry about that.
Mina (2018-09-30):
Do you mean by "look like music" to just follow music theory rules or
actually sound like real music.
Just curious.
In my mind, the purpose is to have a signal that can exercise filters
and codecs, trigger as many code paths as possible, without requiring an
external sample.
Therefore, it must "look like music" from the point of view of analysis
done by filters and codecs. And since some of the have quite subtle
psychoacoustic models tweaked for music, following music theory rules is
probably a requirement.
Sounding like real music seems much more difficult, and possibly not
that useful. But I think that if the code is clean, if somebody comes up
with a better algorithm to sound more catchy, it can be added in the
same filter as an option.
And, to answer Carl Eugen's question: No, I do not have a student lined
for this project.
Regards,
I think I can start on that, I was already searching for my next
contribution. Pedro's reply seems pretty useful.
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
