GUIDE

Unreal Engine text to speech: AI voice lines for your game

An Unreal project collects hundreds of voice lines long before a cast is booked: barks, quest dialogue, tutorials and placeholder scenes that change every week. Text to speech gives each line a voice the day it is written and redoes it in seconds when the script changes.

The free SuperClonica engine voices lines offline on an ordinary processor. This guide shows how to generate them in bulk, bring them into Unreal and when runtime speech makes sense.

Browse voices Install the SuperClonica engine

Two ways to give an Unreal game a voice

A player’s PC will not run a speech server, so a packaged single-player game should ship pre-generated lines.

  • Pre-generated lines: WAV files imported as Sound Wave assets, packaged with the game like any recorded dialogue.
  • Runtime speech: the game asks a SuperClonica server for audio, which fits dev tools and online games with their own server.
  • Either way, voicing runs on an ordinary processor, with no fee per line.
  • The model speaks 31 languages; Chinese is not supported yet.

Pick voices free online

On the page of any voice, type a line of your NPC, 30 to 100 characters, choose the language and press Voice it. In a few seconds you hear whether the voice fits the character, with no account.

Generate the lines in bulk

The same text, voice and seed always give the same audio, so regenerate only the lines that changed and keep the script, not the audio, in version control.

  • Install the engine: python -m pip install superclonica-cpu.
  • Keep the lines in a table with an id, a speaker, a language and the text, for example exported from your dialogue spreadsheet.
  • A short script voices each row: superclonica --text "Halt! State your business." --voice guard.json --lang en --output VO_Guard_Greeting_01.wav.
  • Name files by line id and speaker, so your dialogue system finds them and re-imports replace the right assets.

Import into Unreal

The engine writes 16-bit mono WAV, which Unreal imports as Sound Wave assets; drag the files into the Content Browser or re-import a folder after each run. Use Dialogue Wave assets if you localize lines per speaker. For runtime speech, start the server with docker run -d -p 8000:8000 -v superclonica-models:/models superclonica/superclonica:cpu; calling it and playing the returned audio needs C++, such as the HTTP module and a procedural sound wave, or a plugin. Never ship an API key inside a packaged game: let your own game server call the speech server.

Voices you can ship

For prototypes, any catalog voice is fine. For a released game, use a voice you have the rights to: the ten stock voices of the model, M1 to M5 and F1 to F5, or your own voice trained in the SuperClonica studio from a clean 5 to 10 second recording of an actor who agreed, on a computer with an NVIDIA graphics card, or a custom voice ordered on the exchange. Fan-made character voices are for parody and fan projects, not for commercial games.