Restoring identity through voice in Africa: Senses Hub x ElevenLabs

Millions of people around the world live with speech impairments or loss of voice caused by stroke, ALS, cerebral palsy, cancer, or injury. For many, technology provides a bridge—Augmentative and Alternative Communication (AAC) devices that turn text into speech.

Built primarily on American and British English models, most AAC voices sound foreign, robotic, or disconnected. They don’t reflect the accents, tones, or cultural context of users in places like Nairobi, Lagos, or São Paulo. The result is more than a technical gap; it’s a loss of personal identity and social belonging.

When your voice doesn’t sound like you, communication becomes harder. Misunderstandings increase, confidence drops, and stigma deepens. For too many, every conversation becomes a reminder that their voice isn’t their own.

Giving people their voices back

Senses Hub, based in Nairobi, is one of Africa’s leading accessibility and assistive technology research centers. The organization works to make technology inclusive, affordable, and relevant to local communities.

Together with the ElevenLabs Impact Program, Senses Hub is developing localized voice models for AAC systems across Africa, creating voices that sound authentic, natural, and human.

ElevenLabs’ deep learning models reproduce speech with high emotional accuracy and multilingual flexibility. With only a few seconds of recorded audio, we can restore a person’s original voice or create a new one that feels personal and culturally aligned.

For those who don't have recordings, a voice donor system allows volunteers to contribute their voices. These recordings are then adapted to reflect the user’s preferences and identity, ensuring everyone can sound like themselves.

Building an inclusive voice ecosystem

The project begins in East Africa—Kenya, Uganda, Tanzania, and Rwanda—and expands through Senses Hub’s regional partner network. It focuses on two groups:

• People who can recreate their natural voices from existing recordings.
• People without prior speech who can select or adapt donor voices.

These voices will integrate into AAC tools, communication apps, and assistive platforms, improving day-to-day communication in schools, workplaces, hospitals, and homes.

How the technology works

ElevenLabs’ Text to Speech and Professional Voice Cloning technology deliver:

• High-fidelity voice recreation from minimal audio samples.
• Accent and language flexibility that reflects local diversity.
• Expressive, human-like delivery that captures tone and emotion.
• Simple API integration for AAC developers and platforms.
• Ethical voice creation, based on consent, privacy, and transparent data use.

Real-world impact

A stroke survivor regains the ability to speak in their own voice.

A student with cerebral palsy contributes confidently in class.

A professional with ALS conducts meetings using a natural, regionally familiar voice.

Across healthcare, education, public services, and daily life, this collaboration aims to replace generic synthetic speech with voices that restore confidence and dignity.

Making an impact in Africa

Through our partnership with Senses Hub, we’re helping to build the foundation for a more inclusive voice technology ecosystem in Africa. By combining global innovation with local expertise, we’re creating solutions that reflect the diversity, culture, and identity of the communities they serve.

Join the initiative

If you’re a nonprofit organization working to expand accessibility through voice, we invite you to collaborate with us by applying to our Impact Program.