Closed stephband closed 1 month ago
Update! Ah, I see the problem, you need to test on labs.cruncher.ch/scribe/
. Right, instead of putting test audio in this repo, you should be able to use any of the samples you find here:
https://sound.stephen.band/samples/
Such as this URL, for example:
https://sound.stephen.band/samples/steinway-piano/m4a/65-piano-ff.m4a
If they don't work, we need to ask Marco about the CORS permissions these are served with, but I think they are ok.
@halitcelik –
I have removed the
player/...m4a
test audio file from the repo, and added*.m4a
to.gitignore
. You will have to put the file back in locally, but please pull and make sure your.gitignore
has*.m4a
listed first.The problem with committing audio is that they are big files, and as the git repo carries around a history of everything that has ever been in it, they can make the repo very big.
I want to solve this problem by uploading a load of samples to S3 or another server, with some very permissive CORS settings. That's what @mbi and I were chatting about today. You could look into that next.
This repo contains orchestral samples in
.wav
format: https://github.com/soundio/VSCO-2-CE. I want to upload them all to S3, or to sound.stephen.band, ideally preserving the directory structure, along with.m4a/256kbit
encoded versions of the same files in the same locations. Why two copies? The compressed.m4a
files will be for quick download, but I want to preserve the ability to switch back to high quality.wav
files when needed.