> ## Documentation Index
> Fetch the complete documentation index at: https://docs.runable.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Voice Swap

> Replace the voice in an existing recording while preserving the original emotion, pacing, and timing.

Voice swap takes an existing audio recording and replaces the voice with a different one. The original emotion, pacing, and timing are preserved. Only the voice identity changes.

## How to swap a voice

<Steps>
  <Step title="Upload your audio file">
    Attach the audio recording you want to re-voice using the **+** button in the input bar.
  </Step>

  <Step title="Tell the agent which voice to use">
    Specify a preset voice or a cloned voice ID.

    <Prompt description="Swap to a preset voice.">
      Swap the voice in this recording to Sarah's voice. Keep the same emotion and timing.
    </Prompt>

    <Prompt description="Swap to a cloned voice.">
      Re-voice this recording using my cloned voice.
    </Prompt>
  </Step>

  <Step title="Review and download">
    The agent generates a new audio file with the replaced voice. The pacing, emotion, and delivery match the original. Download the result as MP3.

    <Frame caption="The original and re-voiced audio files side by side in the chat.">
      <img src="https://mintlify.s3.us-west-1.amazonaws.com/runableinc/images/placeholder.png" alt="Chat showing two audio files — the original recording and the re-voiced version with the new voice" />
    </Frame>
  </Step>
</Steps>

## When to use this

* Re-recording narration in a different voice without re-performing the delivery.
* Applying a cloned voice to content that was recorded by someone else.
* Creating multiple versions of the same content in different voices.
* Anonymizing a voice while keeping the natural delivery.

## What you cannot do

* You cannot swap voices in a video file directly. Extract the audio first, swap the voice, then recombine in an external editor.
* You cannot swap only parts of a recording. The entire file is re-voiced.
* You cannot adjust the emotion or pacing. The swap preserves the original delivery exactly.

## Next steps

<CardGroup cols={2}>
  <Card title="Dubbing" icon="language" href="/dubbing">
    Translate and dub audio into another language.
  </Card>

  <Card title="Voice Cloning" icon="clone" href="/voice-cloning">
    Clone a voice to use for swaps.
  </Card>
</CardGroup>