Acapela Group
Belgian synthetic voice specialist known for custom brand voices and for building personal voices for people losing the ability to speak
Quick Overview
| Company | Acapela Group (part of Tobii Dynavox) |
|---|---|
| Category | AI Voice & Speech |
| Headquarters | Mons, Belgium |
| Founded | 2003 |
| EU Presence | Yes - Belgium (EU) |
| Data Location | EU; embedded and offline deployment available |
| Open Source | No |
| GDPR Compliant | Yes |
| Self-Hosting | Yes - embedded SDK for offline devices |
| Main Features | Text-to-speech in 30+ languages, custom brand voices, my-own-voice personal banking, embedded SDK, child and character voices |
| Pricing | From about EUR 5/month (consumer apps) / custom pricing for embedded and enterprise |
| Best For | Assistive technology, embedded devices and organisations wanting a distinctive brand voice |
| Replaces | ElevenLabs, Amazon Polly, Nuance Vocalizer |
Detailed Review
Acapela sits in an unusual position in the synthetic voice market. Commercially it builds custom brand voices and supplies embedded speech to device manufacturers. Socially, its my-own-voice programme lets people with degenerative conditions such as ALS bank their voice while they still can, so that a communication device can later speak in something recognisably theirs.
The company is based in Mons, Belgium, and is now part of Tobii Dynavox, the Swedish assistive technology group — so the ownership as well as the operation stays European.
Voice Banking and Assistive Use
my-own-voice takes a set of recordings from someone who still has speech and produces a synthetic version they can use after they lose it. For families facing a diagnosis of ALS or a laryngectomy, the difference between a generic device voice and one built from the person's own recordings is not a technical detail.
This work also explains Acapela's engineering priorities: intelligibility at speed, reliable pronunciation and offline operation matter more in assistive contexts than dramatic expressiveness.
Embedded and Offline Deployment
The embedded SDK runs entirely on-device, with no network dependency. Communication aids, in-car systems, public transport announcements and industrial equipment all need speech that works when connectivity does not, and that requirement rules out most of the current generation of cloud-only AI voice services.
For data protection, on-device synthesis means no text is transmitted anywhere — the cleanest possible position for medical or personal communication.
Character and Child Voices
Acapela maintains an unusually varied voice catalogue, including child voices and stylised character voices that are hard to source elsewhere with proper commercial licensing. For educational products, games and children's media, that catalogue is a practical differentiator.
Limitations to Consider
Acapela's voices are built for clarity and reliability rather than for the emotional range of the newest generative models, so for cinematic or highly expressive work Respeecher or a generative provider will produce better results. Pricing for embedded and enterprise use is quoted rather than published, and the self-serve options are aimed at consumer apps rather than professional production.
Who Should Use Acapela Group
Assistive technology providers, device manufacturers, educational publishers and organisations that need a distinctive owned brand voice with proper licensing. It is also the first place to look for anyone considering voice banking ahead of losing their speech.
Pros and Cons
Pros
- my-own-voice banking for assistive use
- Fully offline embedded SDK
- Unusual catalogue including child voices
- European ownership under Tobii Dynavox
- 30+ languages with reliable pronunciation
Cons
- Less expressive than generative rivals
- Enterprise pricing not published
- Consumer offering is limited
Alternatives to Acapela Group
Other European ai voice & speech worth comparing before you decide:
Synthesia
United Kingdom · AI video avatars with synthetic voice in 140+ languages
Respeecher
Ukraine · Speech-to-speech voice cloning used in film and television
ReadSpeaker
Netherlands · Enterprise text-to-speech in more than 50 languages
CereProc
United Kingdom · Character voices and voice reconstruction from Edinburgh
See all European ai voice & speech or compare them directly against ElevenLabs on our alternatives page.
Frequently Asked Questions
It is Acapela's voice banking service. Someone who still has speech records a set of prompts, and Acapela builds a synthetic voice from them that can be used on a communication device later. It is aimed at people facing conditions such as ALS or surgery affecting the voice.
Yes. The embedded SDK runs entirely on-device with no network dependency, which is why it is used in communication aids, vehicles and public announcement systems.
Yes. It is based in Mons, Belgium, and is part of Tobii Dynavox, a Swedish assistive technology group, so both operation and ownership remain in the EU/EEA.
Yes, custom voice creation is one of its commercial services, with licensing suited to organisations that want a distinctive voice they own across products and channels.