change_voice converts an existing audio clip's voice to a different Cartesia voice while keeping the original delivery and timing intact — it's the tool for when the PERFORMANCE is right but the voice itself needs to change, rather than re-typing and re-generating from scratch.
Because it operates on audio (not a video file directly), the typical flow for changing a voice IN a video is: get the video's dialogue as a standalone audio clip, run change_voice on it, then reattach with attach_audio_to_video.
Powered by
Convert an existing audio clip's voice to a different Cartesia voice, keeping the delivery/timing (Cartesia voice changer). Use when the user has an audio recording and wants it re-voiced rather than re-typed and re-spoken.
What to tell the agent
Versely's agent maps this plain-English request directly onto change_voice. You don't need to know the parameter names — just describe what you want.
“Change this audio clip's voice to a different Cartesia voice, keep the exact same timing and delivery.”
How it works
1. Isolate the audio
Start from the standalone audio_url of the recording you want re-voiced — pulled from the video's audio track.
2. Pick a target voice_id
Choose from Cartesia's catalog, or a voice you cloned yourself with clone_voice_from_audio.
3. Set the output format
output_format controls the returned audio's encoding.
4. Reattach to the video
Use attach_audio_to_video (mode 'replace') to put the re-voiced track back onto the original footage.
What it costs
Billed per audio clip processed through change_voice, plus the flat attach_audio_to_video fee to put it back on the video. Check estimate_cost for the exact number.
The formula behind that number: What does an AI video with sound cost? — flips the per-second rate with an audio toggle rather than adding a separate fee.
Limits & things to know
- This re-voices an existing recording's DELIVERY — it doesn't let you type new words; the words spoken stay the same, only the voice changes.
- Operates on audio, not video directly — you'll pair it with attach_audio_to_video to get the new voice back onto footage.
- Only available for Cartesia voice targets, including your own cloned voices.
Who uses this
- Anonymizing a speaker's identity while keeping their exact delivery
- Swapping a placeholder voice for the final chosen one
- Correcting a recording made with the wrong voice profile
- Creative voice-swap effects
Frequently asked questions
Does this let me change what's being said?+
No — change_voice keeps the original delivery and timing, and only swaps the voice itself. If you need different words entirely, generate new audio with generate_speech instead.
Can I change a video's voice directly?+
change_voice operates on an audio file. To change a video's voice, extract or reference the dialogue as audio, run change_voice on it, then reattach the result with attach_audio_to_video.
Can I use my own cloned voice as the target?+
Yes — any voice_id from clone_voice_from_audio works as a target for change_voice.
Related Versely tools
Related editing jobs
Change the Voice in a Video inside Versely
The all-in-one AI studio for creators. 60+ models for video, image, voice, music and lipsync in a single app.