Analyze a recording
Choose a file, check it by ear, then count the sources and, when the count is low enough, separate them.
No audio handy? Make a synthetic mixture to test the pipeline.
Uploaded audio is processed on the server in an isolated job folder and deleted automatically after 30 minutes and on server startup.
No result yet
Add a recording on the left. The estimated count, its confidence and the window-by-window votes will appear here.