I record a lot of scripted material, and I often make mistakes. When I do I pause and re-read the mistaken phrase: "Two households, both a— both alike in dignity, in fancy — fair Venice — Veronica — in fair Verona, where we lay our scene" I use Descript's Import feature to match the recording to the text, and then look through the waveforms for mismatches — but since Descript can recognize where the recording varies from the script, can't it also recognize these repeats? 99% of the editing work I do on the text of a scripted recording involves re-takes, and in 99% I have to delete the first utterances of a phrase or subphrase and keep the last one: "Two households, [both a—] both alike in dignity, [in fancy —] [fair Venice —] [Veronica —] in fair Verona, where we lay our scene" If Descript were able to recognize when phrases in a script were repeated (or flubbed) before they were finally gotten right, and delete those flubs, editing recorded scripted material in Descript would become a 1-step operation!
