From first take to final cut

Underlord

Underlord is Descript's AI co-editor that turns a raw recording into an edited video from a text instruction. It can cut filler, pick the best takes, add scenes, and generate clips, and you can review or undo every change.

Text-based editing

Text-based editing lets you edit video and audio by editing the transcript. Deleting a word removes that moment of media; moving a sentence moves the footage with it.

Studio Sound

Studio Sound is an AI audio filter that removes background noise and echo and enhances the speaking voice. It applies in one click to any track.

Descript MCP

Descript MCP is a connector that lets AI agents such as Claude operate Descript to transcribe, edit, and create clips. It runs on the same API available to developers.

Rooms

Rooms is Descript's remote recording studio, capturing up to 4K with a separate track per participant. Guests join from a browser link with nothing to install.

Screen recording

Screen recording captures your screen and camera together in one session. The recording lands in the editor already transcribed.

Transcription

Transcription converts audio and video to accurate, timestamped text in dozens of languages. The transcript is the editing surface, not just an export.

Remove Filler Words

Remove Filler Words finds and deletes ums, uhs, likes, and long pauses across a whole project in one pass. You can review each cut before applying it.

Remove Retakes

Remove Retakes detects repeated attempts at the same line and keeps the last good one. Record as many takes as you need and clean them up afterward.

Regenerate

Regenerate repairs damaged or mismatched audio by generating replacement speech that matches the surrounding tone. Video Regenerate does the same for picture, adjusting mouth movements to match a typed correction.

Eye Contact

Eye Contact is an AI effect that redirects your gaze toward the camera when you were reading off-screen. It works on existing footage, no teleprompter required.

Green Screen

Green Screen removes or replaces the background behind a speaker without a physical green screen. Drop in any image or video instead.

Captions

Captions generates styled, editable subtitles from the transcript in one click. Style them to match your brand or burn them in on export.

Create Clips

Create Clips mines a long video or podcast for its strongest moments and reformats them for social platforms. Each clip stays editable in the project it came from.

Translation

Translation dubs and subtitles a video into dozens of languages from the original recording. The translated audio keeps the speaker's voice.

Templates

Templates are prebuilt layouts and workflows you start from instead of an empty timeline. Pick one, drop in your footage, and the structure is already there.

Integrations

Integrations connect Descript to the tools around it — publish straight to YouTube, sync with Google Drive, or round-trip through Adobe.

Podcasting

Podcasting in Descript covers the full workflow — multitrack recording, text-based editing, audio repair, and publishing — for audio or video shows.

Import

Import brings existing video, audio, or a Zoom recording into a project, transcribing it on arrival. There is no separate page for import; it is the first step of every other feature here.

Export

Export writes your finished project to any common format and resolution, or publishes it directly to a connected platform.

Scenes and layouts

Scenes and layouts arrange what is on screen at any moment — speaker framing, screen share, B-roll, and titles — without touching a timeline.

Collaboration

Collaboration lets several people comment on and edit the same project, with comments anchored to the moment in the video they refer to.