musikinformatik / SuperDirt

Tidal Audio Engine
GNU General Public License v2.0
524 stars 76 forks source link

Making use of audio file metadata #122

Open yaxu opened 5 years ago

yaxu commented 5 years ago

It could be nice to be able to make use of audio file metadata such as cue points.

These can be embedded in wav files.. It seems this isn't generally accessible to SC but might be on the way: https://scsynth.org/t/reading-file-markers/241/6

We have talked about making use of off-line analysis, such as onset detection to index inside audio files, in the past.

telephon commented 5 years ago

If you export markers from audacity, you could already use those. It is just a little unclear how the parameters would look like.

jkbd commented 5 years ago

Replay Gain would also be interesting to have. One could use it to predefine amp settings for a loudness balanced sample set.

yaxu commented 5 years ago

In the case of audacity regions, I suppose it would really be tidal rather than supercollider that needs this info, as the main event scheduler is there.. As you might want to set a sound in time relative to its middle or end

jkbd commented 5 years ago

you might want to set a sound in time relative to its middle

Yes! I always missed this feature with samplers, for example to align a cymbal swell crescendo to a beat.

yaxu commented 5 years ago

Here in section 3 Bel refers to points (rather than regions) in a sound object as pivots, after Duthen and Stroppa, https://hal.archives-ouvertes.fr/hal-00134179/document