
Type something to start
How this template works
The voice generator reads your line in the character voice you pick, with the kind of delivery dialogue needs.
1Open Audio GenerationStep 1 · Audio GenerationVoice the line
The expressive speech model voices your line in the chosen character's voice and returns an MP3.
Voice Expressive · MP3 · 1 credit per 100 characters
Tips for the best result
- Write it the way it's said: punctuation shapes the delivery.
- Short lines (one or two sentences) sound most natural.
- Spell unusual names the way they sound.
Do it yourself for more control
- Open Audio Generation for more voices, languages and delivery controls.
- Add [whispers], [laughs] or [sighs] in the line for performance cues.
- Keep one voice per character across your whole script.
Good to know
Voices are presets, not clones; for a specific actor's voice, record or license it.
Made with this template
The card in the corner shows what went in.




Each example is an actual run of this template at Standard quality, made from a typed idea.
Run this template from Claude, ChatGPT or Cursor
Connect the 3D AI Studio MCP to your AI assistant and paste the brief into the chat. It runs the same steps with the same prompts and settings, shows you the result in the conversation and saves it to your library.
- Connect the MCP once in your account settings. It's included in every paid plan.
- Copy the brief and swap the example in the first line for your own.
- Paste it into a new chat. Your assistant runs every step and shows you each result.
generate_speechget_generationshow_generationUse the 3D AI Studio MCP to run the "NPC Voice Lines" template (https://www.3daistudio.com/templates/npc-voice-generator) for me.
My line: Ah, a traveller! The road north is dangerous, take this map.
Voice: Wise elder (voice George)
Follow these steps in order. The prompts and settings are the template's own, so use them word for word and only put my line where it says {MY LINE}.
1. Voice the line
Call generate_speech with voice "George", output_format "mp3" and text {MY LINE}, spoken exactly as I wrote it.
After each generation call, keep polling get_generation with its task_id until it completes, then show me the result with show_generation before you start the next step.Similar templates
More templates like NPC Voice Lines, ready to use the same way.







