List view
The representation of a diff using our current experimental Python frontend shows artifacts from chopping up the audio into shift-resistant chunks. (An insertion does not often occur exactly at a chunk boundary!) To fix this, once we've got the general set of diff hunks, we need to scan each hunk more precisely to know exactly when it starts and stops. This milestone should also include the end-to-end adiff testing.
No due date•10/10 issues closed* CLI * Visual representation of diff * Think about audio playback of diff
Overdue by 11 year(s)•Due by January 18, 2015•3/3 issues closed