Intron Health, a clinical speech recognition startup, has raised $1.6 million (N2.55 billion) in pre-seed funding to advance the development and deployment of its innovative speech recognition tool, designed to understand African accents.
The funding round was led by Microtraction, with significant participation from Plug and Play Ventures, Jaza Rift Ventures, Octopus Ventures, Africa Health Ventures, OpenseedVC, Pi Campus, Alumni Angel, Baker Bridge Capital, and several angel investors.
With the new funding, Intron Health said it would explore further advancements, including perfecting noise cancellation, ensuring platform functionality in low bandwidth environments, enabling the transcription of multi-speaker conversations, and integrating text-to-speech capabilities.
The startup
Intron Health operates in a competitive market alongside established players like Helium Health and other health tech startups aiming to address Africa’s healthcare challenges.
The CEO and founder, Tobi Olatunji, highlights that the company boasts Africa’s most extensive clinical speech database. The algorithm is trained on 3.5 million audio clips (16,000 hours) from over 18,000 contributors, primarily healthcare practitioners, representing 29 countries and 288 accents.
“Because we’ve already trained on many African accents, it’s very likely that the baseline performance of their access will be much better than any other service they use,” Olatunji stated. He added that data from Ghana, Uganda, and South Africa is expanding, and the startup is confident about deploying its model in these regions.
Initially, Intron Health wanted to digitize hospital operations in Africa through an Electronic Medical Record (EMR) System.
However, adoption proved challenging as physicians preferred writing to typing. This prompted Olatunji to develop a solution that addressed this fundamental problem: enhancing physicians’ basic data entry and writing tasks.
The company explored third-party solutions for automating tasks like note-taking but encountered numerous issues with constant mistranscription due to thick African accents and the pronunciation of complex medical terms.
This realization spurred the creation of Intron Health’s speech recognition technology, capable of recognizing African accents and integrating them with existing EMRs.
The tool is now used in 30 hospitals across five markets, including Kenya and Nigeria. In one notable case, Intron Health reduced the waiting time for radiology results at one of West Africa’s largest hospitals from 48 hours to 20 minutes.
What you should know
Beyond developing voice technologies, Intron Health is also contributing to speech research in Africa.
The company has recently partnered with Google Research, the Bill & Melinda Gates Foundation, and Digital Square at PATH to evaluate popular Large Language Models (LLMs) such as OpenAI’s GPT-4, Google’s Gemini, and Anthropic’s Claude across 15 countries.
This initiative aims to identify strengths, weaknesses, and potential biases in LLMs to ensure culturally attuned models are available for African clinics and hospitals.
The startup also plans to add intelligent systems for tasks such as prescriptions and lab tests, aiming to reduce doctor errors and enhance patient care.
Intron Health is part of a growing number of generative AI startups in the medical field, including Microsoft’s DAX Express, which is reducing administrative tasks for clinicians.
Leave a Comment