To turn a transcript into a video, paste the strongest passage from the talk, meeting, interview or lecture into the box below, then pick a voice, a visual style and a length in Faceless. It writes a clean script, an AI voice narrates it with captions and a picture per scene, and the video costs 20 credits.
Last updated by the Faceless team
A transcript holds the substance of a talk without the camera, which makes it good raw material for a new video. A one hour interview runs about 9,000 words, so look for the two or three moments worth retelling, each 75 to 225 spoken words. Faceless does not use the original recording or the speakers' voices: an AI voice narrates a new script written from the passage you paste, with a generated picture for each scene and captions timed to every word. Cut the fillers, crosstalk and timestamps first. A video in the Storyboard style costs 20 credits, and plans start at $29/month on the Starter plan.
Skim for a story, a clear explanation or a strong opinion. A one hour transcript usually holds three to five such passages, each one a video of 30 to 90 seconds.
Delete timestamps, speaker labels, "um", "you know", false starts and crosstalk. Keep the speaker's point. The box takes up to 800 characters, about 130 words.
Choose a visual style, a narrator voice, the language and under 30, 30 to 60 or 60 to 90 seconds. The narrator is an AI voice, not the person in the recording.
Turn on Approve each script first on the Posting step. Faceless drafts the script ahead of time, and you can paste in the speaker's exact sentences with attribution, such as "As she put it,". Nothing is charged until you approve.
Check names and numbers in the editor, then download or post to a connected YouTube, TikTok or Instagram account. It is set up as a series that can keep making videos on the topic on a schedule, or you can pause it after one.
Find answers to common questions about our platform
Start creating amazing AI-powered faceless videos in minutes with Faceless