CereProc
Edinburgh speech-synthesis company known for expressive TTS and voice cloning.
About
CereProc is a speech synthesis company based in Edinburgh, Scotland, founded in 2005. It specializes in natural, expressive text-to-speech voices, regional-accent voices, and voice cloning. Its CereVoice engine runs across mobile, embedded, desktop, and server platforms, and the company holds a patent on emotional speech synthesis.
CereProc voices are built with both unit-selection synthesis (from large recorded speech databases) and parametric statistical modelling, the latter enabling private voice cloning from far less recorded speech. The company famously recreated film critic Roger Ebert's voice from interviews and recordings after illness left him mute, and collaborated with Accenture on an open-source non-binary voice.
Practical notes: CereProc is a commercial, enterprise-oriented vendor rather than a self-serve SaaS; custom voice builds are commissioned projects. Coverage of the company is thinner than for newer AI voice startups, so current product details should be confirmed on cereproc.com.
What it can do
- Expressive text-to-speech in 20+ languages
- Regional-accent voice portfolio
- Private voice cloning from limited recordings
- Emotional speech synthesis (patented)
- Embedded, mobile, desktop, and server deployment
Setup
Contact CereProc for commercial voice licensing or custom voice builds via cereproc.com.
Prerequisites
A commercial use case; custom voices require recorded speech from the voice donor.
Use cases
Works with
Trust score — 6/10
Scored 6/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 87; Trending score 58/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.