- What can AI text-to-speech be used for?
- Narration for video and tutorials, audiobook and long-form reading, e-learning modules, dubbing into another language, IVR and phone prompts, and accessibility voiceovers. The Speech tab of the Audio Workshop turns a script into an audio file you can download.
- How is text-to-speech priced?
- Per 100 characters of script rather than per generation, so the length of your text is the main cost driver, not which model you pick. A short caption costs very little; a full audiobook chapter costs proportionally more.
- How do I choose a voice?
- Voice matters more than model here. The Speech tab offers a system voice library you can preview, and choosing a BytePlus voice automatically routes generation to the BytePlus engine. You can also train and use a cloned voice once you have one.
- Can I clone my own voice?
- Yes, through the voice cloning tool in the Audio Workshop. Once a cloned voice finishes training it becomes selectable for text-to-speech alongside the system voices.
- What languages are supported?
- The MiniMax speech models support a wide range of languages and offer a language boost setting to improve pronunciation for a specific one. Check the individual model page for its current language list.
- Can I use AI voiceovers commercially?
- As with the other model types, that is set by the provider's terms for the model and voice you used. Cloned voices carry the additional requirement that you have the right to clone the voice in question.