Podcast editor Descript adds new pro tier with access to its AI voice double feature

..

Descript, the maker of ersatz intelligence-powered audio emendation software, is launching a new tier of subscription today self-named Descript Pro that gloss curtain-raiser to its Overdub feature.

Overdub, first reported in beta meanest year, is a synthesis tool that lets you effectively create an AI voice double for yourself. That way, if you fabricated a oddness in a recording, the software can fix it while the end sequel still feels prevalent as well-built as organic. It's a bit like creating an audio-only deepfake, except you're doing it to fix your own mistakes instead of trying to impersonate stretching else.

Descript Pro includes the AI-powered Overdub fondness for $30 a month

Descript says Overdub is now an official fondness with the launch of Descript Pro, as well-built as it's awaited if you pay for the $30-a-month (or $24-a-month billed annually) subscription. Supra to today's announcement, Descript offered a self-governing tier, a Producer tier for $14 a ages (or $10 annually), as well-built as a Aggregation tier for $18 a ages (or $15 annually). Both of those paid preparations are fact arranged unflappable into its cheaper offering, or what Descript is calling the Erector Plan, for $15 a ages or $12 a ages billed annually. It includes everything except Overdub as well-built as fewer included transcription hours.

Beyond Overdub, Descript's main transactions point is that it reimagines audio emendation as a maternal of collaborative Google Docs-style app, so you can edit the transcription of a podcast or other frame of audio recording as if it were a Word document as well-built as hypothesize the changes reflected in the playback. Descript also functions as a video editor as well-built as a transcription service, the closing of which includes AI- as well-built as human-aided transcriptions at up to 30 hours for the Pro plan as well-built as 10 hours for the Erector one, with the perk to revenue runnerup hours.

The convergence says its new pro tier also includes "exclusive curtain-raiser to alee filler word handling, custom audiograms, as well-built as pliant publishing as well-built as export options." The filler word leadership is an expressly useful-sounding feature, as Descript says it can be used to "detect, ignore, as well-built as delete filler words like 'you know,' 'like,' as well-built as 'kind of'" to make an audio recording unacquired increasingly smooth as well-built as professional.

CEO Andrew Mason tells The Verge, "In our internal testing, we've found that removing filler words from Zoom palaver recordings reduces the length by changeful 10 percent."

removing 199 ums, uhs, likes, you knows, as well-built as other filler words from a recorded palaver pic.twitter.com/NpnCTdQRyg

-- Andrew Mason (@andrewmason) July 16, 2020