AI Speech to Signs: Wins Top Prize

In a significant leap forward for assistive technology, a Kenyan innovator has developed an app that translates spoken language into sign language using AI-powered 3D avatars. Terp 360, created by Elly Savatia, has been awarded the prestigious Africa Prize for Engineering Innovation by the UK’s Royal Academy of Engineering, taking home £50,000 ($67,000) in recognition of its transformative potential for the deaf community.

The Innovation

The app functions as a real-time translation service, converting speech and text into sign language performed by photorealistic avatars. As Savatia explains, “Think of it as Google Translate for sign language, that takes speech and text input and outputs photorealistic avatars signing.” This innovative approach addresses a significant communication barrier faced by the deaf community in Kenya and potentially across Africa.

What makes Terp 360 particularly noteworthy is its culturally relevant approach to sign language translation. Unlike existing technologies primarily designed for Western sign languages, this app was developed in close collaboration with deaf and hard-of-hearing Kenyans. The development team recorded over 2,300 signs, including commonly used phrases and words, using motion sensors attached to signers’ hands to capture precise movements with remarkable accuracy.

Addressing Accessibility Challenges

The need for such technology in Africa is profound. According to the World Health Organization, over 80% of people with disabilities in low-income countries lack access to basic assistive technologies¹. In Kenya specifically, despite government legislation requiring employers to reserve at least 5% of jobs for people with disabilities, Savatia notes a persistent gap: “Companies cannot afford interpreters… [and] they just don’t have the tools to effectively integrate these people.”

Savatia further explains the daily struggles faced by the deaf community: “To go to the workplace, education, health care, you have to communicate. But the deaf community, they’re left behind.” The shortage of human sign language interpreters and their high cost create significant barriers to accessing essential services.

Current Capabilities and Future Expansion

Currently translates from English and Swahili into Kenyan Sign Language
Plans to expand to include Rwandan, Ugandan, South African, British and American sign languages by mid-2027
Uses motion capture technology to record and learn up to 1,000 words per day
Addresses one of the estimated 30 sign languages used across Africa²

This expansion approach recognizes the diversity of sign languages across the continent. There are more than 300 sign languages used worldwide, and around 30 in Africa, each with their own unique grammar, vocabulary, and cultural context. The app’s developers plan to partner with local NGOs and media organizations that have existing sign language video datasets to train the AI in these additional languages.

Recognition and Impact

The recognition from the UK’s Royal Academy of Engineering highlights not just the technical innovation but also the social impact of the solution. Rebecca Enonchong, chair of the prize-judging panel, emphasized this aspect: “One of the criteria that we use to judge is social impact. We want to know that there is an impact that goes beyond just the technology.” She further noted, “We want to know that there is an impact beyond creating jobs, creating wealth, that the solution brings. And assistive technologies are one of those areas that are underserved… not just on the continent, but in the world.”

The app was selected from four finalists across Africa by a panel of seven judges during a ceremony held in Dakar, Senegal. Enonchong praised Savatia’s work, stating, “What really stood out about Elly’s solution, and Elly himself, is the level of innovation. It was really a demonstration that Africans are able and capable of using cutting-edge technology to solve problems, not just on the continent but beyond.”

Technical Advantages

While other avatar-based sign language translation technologies exist, Savatia points out their limitations: “These are not designed with African sign languages or cultural context in mind.” The development of a dedicated motion capture studio in Nairobi demonstrates a commitment to authentic representation of African sign languages. The studio has the capacity to record and learn 1,000 words a day, allowing for rapid expansion of the app’s vocabulary and language capabilities.

This approach represents a significant advancement in machine learning-based sign language recognition³ by using wearable glove devices that capture intricate hand movements essential for accurate sign language translation.

Broader Implications for African Innovation

This achievement also underscores the growing prominence of Kenyan tech innovation. The country has become a hub for tech startups, with initiatives such as Nairobi’s “iHub” and “NaiLab” providing spaces for innovation and entrepreneurship⁴. Young Kenyan entrepreneurs are increasingly leveraging technology to solve local challenges, from financial inclusion to healthcare access.

As the Kenyan technology ecosystem continues to mature, success stories like Terp 360 demonstrate the potential for homegrown solutions to address global challenges. Kenya’s status as an emerging leader in tech development⁵ is increasingly recognized both regionally and internationally.

Looking Ahead

Beyond its immediate application, Terp 360 represents a model for inclusive technology development. By centering the needs and expertise of the deaf community in its creation, the app demonstrates the importance of participatory design in accessibility solutions. The developers’ approach of partnering with local communities and organizations ensures cultural authenticity while building technical accuracy.

Savatia envisions the app as a scalable solution: “We see ourselves as an enabler. We are able to do sign language, but at scale.” As the technology continues to evolve, it could serve as a foundation for broader digital accessibility initiatives across Africa. With the continent home to diverse linguistic landscapes and growing tech ecosystems, innovations like Terp 360 point toward a future where technology serves all communities equitably.

Conclusion

The success of Terp 360 represents a confluence of technological advancement, social awareness, and cultural sensitivity. By addressing a fundamental accessibility challenge through homegrown innovation, the app not only serves the immediate needs of Kenya’s deaf community but also demonstrates the potential for African-developed solutions to have global impact. As the technology expands to support more languages across the continent, it could serve as a model for other assistive technologies designed with local context and needs at the forefront.

The innovation also highlights a broader narrative about African entrepreneurship and technological capability. As Enonchong noted about Savatia’s solution, “It was really a demonstration that Africans are able and capable of using cutting-edge technology to solve problems, not just on the continent but beyond.” In bridging communication gaps and creating inclusive digital environments, apps like Terp 360 represent the future of accessible technology in Africa and beyond.