Available now

Chatterbox: text to speech

Chatterbox reads up to 5000 characters aloud and can match the style and tone of a reference recording you upload.

Model ID

Input

Fields marked * are required. Review the price before generating.

Example. Edit it or write your own.

Uploading a file needs an account. Create account

MP3, WAV · up to 50 MB

Additional settings

Exaggeration factor for the generated speech (0.0 = no exaggeration, 1.0 = maximum exaggeration).

Temperature for generation (higher = more creative).

Price$0.04 / run

You sign up first. Your settings are kept.

About this model

Overview

Chatterbox turns text into speech. Add a short recording and it follows that speaker's style and tone; without one it uses its own voice. Tags such as <laugh>, <sigh> or <gasp> in the text add those sounds where you place them.

Exaggeration makes the read more dramatic, temperature makes it more varied, and a fixed seed repeats a result.

What you can build

  • Voice lines for memes, shorts and games
  • Narration in the tone of a sample recording
  • Spoken replies for chat agents
  • Dialogue with laughs, sighs and groans

Prompting tips

  • Use a clean reference recording of one speaker with no background music.
  • Keep exaggeration low for narration; raise it for characters and comedy.
  • Place an emotive tag right where the sound belongs: "Oh no <sigh> not again."

What to check

Listen for names, pronunciation, pauses, clipping and the intended tone before publishing.

Inputs

  • Text (required): The text to be converted to speech (maximum 5000 characters). You can additionally add the following emotive tags: <laugh>, <chuckle>, <sigh>, <cough>, <sniffle>, <groan>, <yawn>, <gasp>.
  • Audio (optional): Optional URL to an audio file to use as a reference for the generated speech. If provided, the model will try to match the style and tone of the reference audio.
  • Exaggeration (optional): Exaggeration factor for the generated speech (0.0 = no exaggeration, 1.0 = maximum exaggeration).
  • Temperature (optional): Temperature for generation (higher = more creative). From 0.05 to 2.
  • Cfg (optional) From 0.1 to 1.
  • Seed (optional): Useful to control the reproducibility of the generated audio. Assuming all other properties didn't change, a fixed seed should always generate the exact same audio file. Set to 0 for random seed..

Output and limits

  • Sung or spoken words can come out unclear or differ from your text.
  • Exact length, tempo and key are not guaranteed; listen to the whole result before you use it.
  • Prompts that name real artists or protected songs can be refused.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • up to · 1,000 ASCII characters: $0.04 / run