The work
What AI voice work actually involves
AI companies need enormous amounts of recorded human speech to build text-to-speech engines, conversational assistants and cloned voices. That speech has to come from real people - and the good companies want it recorded properly, with rights they can prove.
That is where you come in. Typical projects we recruit for:
- TTS corpus recording - reading scripted lines, consistently, across long paid sessions. The backbone of every synthetic voice.
- Voice cloning with consent - lending your actual voice identity to a product, under a licence that names where it can appear and for how long.
- Conversational and character voices - expressive recording for AI assistants, games and interactive products.
- Multilingual and accent datasets - natural speakers of specific languages, dialects and regional accents. Often no voiceover experience needed.
Some sessions happen remotely from a decent home setup; others are captured at our studios in the UK. Read more about voice acting for AI training or how to sell your voice to AI on ethical terms.
