Topics
recent
AI
Amazon
Image Credits:Intron Health
Apps
Biotech & Health
Climate
Image Credits:Intron Health
Cloud Computing
Commerce
Crypto
endeavor
EVs
Fintech
fund-raise
Gadgets
Gaming
Government & Policy
computer hardware
Layoffs
Media & Entertainment
Meta
Microsoft
Privacy
Robotics
security department
Social
Space
Startups
TikTok
Transportation
speculation
More from TechCrunch
Events
Startup Battlefield
StrictlyVC
Podcasts
picture
Partner Content
TechCrunch Brand Studio
Crunchboard
meet Us
Voice recognition is getting integrated in nearly all facet of forward-looking sustenance , but there remains a magnanimous gap : speaker of minority speech and those with thick accents or delivery disorder like stuttering are typically less able-bodied to use speech - recognition tools that control software , transcribe or automate job , among other routine .
Tobi Olatunji , laminitis and CEO of clinical speech - recognition startupIntron Health , wants to bridge this gap . He take that Intron is Africa ’s prominent clinical oral communication database , with its algorithm train on 3.5 million audio clipping ( 16,000 hours ) from over 18,000 contributors , principally health care practitioners , representing 29 countries and 288 accents . Olatunji says that drawing most of its contributors from the healthcare sphere check that medical full term are judge and becharm correctly for his target market place .
“ Because we ’ve already trained on many African idiom , it ’s very potential that the baseline performance of their access will be much better than any other military service they utilise , ” he said , adding that data from Ghana , Uganda and South Africa is growing and that the startup is positive about deploying the role model there .
Olatunji ’s interest group in wellness technical school stems from two strands of his experience . First , he received grooming and rehearse as a medical Dr. in Nigeria , where he watch at first hand the inefficiencies of the system in that market , let in how much paperwork needed to be occupy out and how intemperately it was to track all of it .
“ When I was a Dr. in Nigeria a pair years ago , even during medical school day and even now , I get irritated easily doing a repetitive undertaking that is not deserve of human efforts , ” he read . “ An easy example is we had to compose a patient role ’s name on every lab order you do . And just something that ’s simple , permit ’s say I ’m picture the patients , and they postulate to get some prescription , they need to get some labs . I have to manually write out every order for them . It ’s just frustrating for me to have to repeat the patient name over and over on each form , the old age , the particular date , and all that . … I ’m always asking , how can we do things well ? How can we make life easier for doctors ? Can we take some tasks aside and offload them to another system of rules so that the doctor can pass their time doing thing that are very valuable ? ”
Those question propelled him to the next stage of his spirit . Olatunji moved to the U.S. to pursue a passkey ’s grade in aesculapian informatics from the University of San Francisco and then another in data processor skill at Georgia Tech .
He then cut his dentition at a number of tech company . As a clinical natural voice communication programing ( NLP ) scientist and researcher at Enlitic , a San Francisco Bay Area company , he built models to automate the extraction of entropy from radiology text reports . He also served Amazon Web Services as a political machine learning scientist . At both Enlitic and Amazon , he focus on natural language processing for healthcare , shaping organisation that enable hospitals to play better .
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
Throughout those experiences , he start out to form estimation around how what was being developed and used in the U.S. could be used to improve healthcare in Nigeria and other emerging markets like it .
The original objective of Intron Health , launched in 2020 , was to digitize infirmary operations in Africa through an electronic medical phonograph recording ( EMR ) system . But take - up was challenging : It plow out physician preferred writing to typewriting , said Olatunji .
That led him to research how to improve that more introductory problem : how to make physicians ’ basic data point incoming , writing , work well . At first the caller front at third - company resolution for automate task such as note - taking and embedding subsist manner of speaking to text applied science into his EMR programme .
There were a lot of offspring , however , because of constant MI - arrangement . It became clear to Olatunji that stocky African accents and the orthoepy of complicated aesculapian terms and names made the borrowing of existing extraneous transcription pecker impractical .
This mark the genesis of Intron Health ’s speech - acknowledgment technology , which can recognize African emphasis and can be integrate with existing EMRs . The tool has to date been take on in 30 hospital across five markets , admit Kenya and Nigeria .
There have been some immediate positive outcomes . In one case , Olatunji said , Intron Health has aid reduce the waiting metre for radiology result at one of West Africa ’s large hospitals from 48 hr to 20 minutes . Such efficiency are vital in healthcare proviso , specially in Africa , where the physician - to - affected role ratio remains one of the lowest in the humanity .
“ infirmary have already spent so much on equipment and engineering … see that they apply these tech is important . We ’re able to bring home the bacon value to help them ameliorate the adoption of the EMR system , ” he tell .
look ahead , the startup is explore unexampled growth frontiers backed by a $ 1.6 million pre - seed round , lead byMicrotraction , with participation from Plug and Play Ventures , Jaza Rift Ventures , Octopus Ventures , Africa Health Ventures , OpenseedVC , Pi Campus , Alumni Angel , BakerBridge Capital and several angel investors .
In terms of engineering , Intron Health is working to hone noise cancelation , as well as ensuring that the platform function well even in low-toned bandwidth . This is in addition to activate the transcription of multi - speaker conversation and integrating textual matter - to - address capabilities .
The plan , Olatunji says , is to add intelligence systems or determination support tools for tasks such as prescription or laboratory tests . These tools , he adds , can help reduce physician errors , ensure adequate patient role care and travel rapidly up their work .
Intron Health is among the develop number of reproductive AI startups in the medical blank space , include Microsoft ’s DAX Express , which are reducing administrative project for clinicians by engender note within seconds . The growth and acceptance of these technology make out as the global speech- and voice - realization market is project to be valued at $ 84.97 billion by 2032 , following a CAGR of 23.7 % from 2024 , harmonise toFortune Business Insights .
Beyond building voice engineering , Intron is also playing a polar role in speech research in Africa , having late partnered with Google Research , the Bill & Melinda Gates Foundation , and Digital Square at PATH to evaluate popular big language models ( LLMs ) such as OpenAI ’s GPT-4o , Google ’s Gemini , and Anthropic ’s Claude across 15 land , to name strengths , weaknesses , and risks of preconception or harm in LLMs . This is all in a bid to ensure that culturally attune simulation are uncommitted for African clinics and hospital .
