Descript: editing podcasts and videos by editing text

The core idea

Descript transcribes your recording, then lets you edit the media by editing the transcript. Cut a rambling sentence with the delete key and that audio disappears from the timeline. For anyone who finds waveform editing intimidating, it's a revelation.

Filler-word removal is the party trick: one command finds every "um" and "uh" and offers to strip them all.

A realistic workflow

Record, import, auto-transcribe, then read the transcript like an editor reads a manuscript. Mark cuts in a first pass, tighten in a second, and only then think about sound polish — leveling voices and reducing room noise with the built-in tools.

Edit for content before editing for sound. Polishing audio you're about to cut is the most common way beginners waste an afternoon.

Where the transcription needs you

Auto-transcription mishears names, jargon and crosstalk, and every transcription error is a potential editing error. Skim the transcript for wrong words in passages you plan to cut — deleting "the wrong bit" is easy when the text lies about what was said.

For multi-speaker recordings, check that speaker labels are correct before relying on them for chapter markers or show notes.

Plans and limits

The free tier includes limited transcription hours per month, with paid plans raising the caps and adding features like higher-quality exports. Check current limits if you're planning a weekly show, since hours run out faster than people expect.