Skip to main content

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 cloning creates a reusable voice profile from one or more audio samples. Once cloned, you can use that voice for text-to-speech, dialogue, and voice swap operations.

​
How to clone a voice

1

Prepare your audio samples

Record or collect audio samples of the voice you want to clone. The better the quality, the better the clone.
Use clean recordings with minimal background noise. Even a single sample works, but more samples improve accuracy.
2

Upload and ask the agent to clone

Attach your audio file(s) using the + button and ask the agent to clone the voice.

Clone a voice from an audio sample.

3

Receive your voice ID

The agent creates the voice clone and returns a voice ID. This ID is reusable across all audio operations in Runable.
Chat showing the agent's response confirming the voice has been cloned with a voice ID returned
4

Use your cloned voice

Reference the cloned voice in any audio request:

Use a cloned voice for text-to-speech.

Voice cloning itself does not cost credits. Credits are only used when you generate audio with the cloned voice (text-to-speech, dialogue, or voice swap).

​
What you cannot do

  • You cannot clone a voice from text. You need at least one audio sample.
  • You cannot edit or fine-tune a cloned voice after creation. Clone again with better samples if the result is not accurate.
  • You cannot use a cloned voice for music generation. Cloned voices work with speech operations only.

​
Next steps

Voice Swap

Replace the voice in an existing recording with your cloned voice.

Text-to-Speech

Use your cloned voice to generate new speech.