How to transcribe a video to text

Attach the video and ask Recut to "transcribe this". It takes out the sound, runs a speech-to-text model on it and gives you the words as text in your project.

Step by step

1. Attach the video or audioAdd the recording to a Recut project. Audio files work too.
2. Ask for a transcriptSay "transcribe this" or "what is said in this video?".
3. Pick the modelRecut suggests speech-to-text models side by side with their prices; transcription usually costs about a cent a minute or less.
4. Run itThe text appears as a result in the project, ready to copy, or to use as the script for captions.

What Recut does

Recut first saves the video's sound as an MP3 on your device, then sends that sound to the transcription model you pick. The models it suggests first are ElevenLabs Scribe, Wizper (a fast Whisper on fal) and Whisper.

The same transcription powers other things in Recut: timing captions to speech, cutting a clip where the words start, and finding a moment in a long video by what is said in it.

Before any transcription runs, the plan shows which model will be used and what it costs. If you have connected your own provider account, such as fal or ElevenLabs, the run is billed there at that provider's price; otherwise it can run on Recut credits.

Tips

• Clear sound gives a better transcript. Clean up a noisy recording first.
• For captions, you don't need a separate transcript: the caption step times the words itself.

What it doesn't do

• The transcript is plain text in Recut. It isn't exported as an SRT or VTT subtitle file.
• Accuracy is the model's, and names and jargon may need correcting.

Related guides

• How to add captions to a video automaticallyAttach a video, type "add captions" and Recut times each word to the speech and burns captions in, styled to read on a phone.
• How to extract the audio from a video (MP3 or WAV)Ask Recut to save a video's sound as an MP3 or WAV. It runs on your device with ffmpeg, free of model costs, and keeps the original video.
• How to turn a long video into a Short or ReelPick the moment by time or by what is said, and Recut cuts it out of the long video, makes it vertical and captions it for Shorts, Reels or TikTok.
• How to dub a video into another language with AIRecut runs an AI dubbing model, such as ElevenLabs dubbing, that translates a video's speech into another language in the same voice.