I edited my first podcast episode by deleting sentences in a text document and it took two hours not ten

I want to write this for people who have wanted to start a podcast but have been put off by the editing step, because Descript changed my assumption about how long that takes.

I recorded my first episode. Fifty-two minutes of raw audio with filler words, long pauses, a few restarts, a section I rambled through and wanted to cut. In a traditional audio editor I would have been in there for the better part of a day, scrubbing through a waveform, cutting clips, trying to make the edits sound seamless.

Descript transcribes the recording automatically. What you see is a text document of everything that was said. To edit the audio you edit the text. Delete a sentence and that sentence is cut from the audio. Move a paragraph to a different position and that audio moves. That is the whole principle and it works.

The Underlord AI suite handles the things you would otherwise do manually. Filler word removal automatically finds and cuts every "um" and "uh". Gap shortening tightens the silences between words and sentences. Audio enhancement cleans up the recording quality. All of those ran in one pass before I even started cutting content.

The Overdub voice cloning is the feature I have not needed yet but understand the value of. If you record a sentence wrong and realise it after the fact, you type the correction and it generates the audio in your voice rather than requiring a re-record.

Dynamic Captions made adding subtitles to the video version straightforward. The stock library covered the intro music I needed.