The innovator section of TikTok strikes again.
How efficient would it be if mathematical instructions could translate different dialects of sign language? A TikToker put it to the test with a quick demonstration of her process.
Under the username @Mrembo, the TikToker shared that she built a machine-learning algorithm to turn Kenyan sign language into audio. Although she had the idea, she was initially unfamiliar with the type of artificial intelligence (AI).
@m.rembo Machine learning Kenyan Sign Language translator #techbuilds #barbthebuilder👷🏾♀️👩🏽🔧 #techkenya #techinafrica #tiktokkenya🇰🇪 ♬ original sound – Mrembo
“Now, first things first, I had no idea about anything [about] machine learning,” she shared. “When I thought about machine learning, I thought about ‘I, Robot’ and Will Smith.”
However, with the help of YouTube and a machine learning course, @Mrembo was caught up to speed on building out her machine learning model.
“We were able to get our hands to get recognized and finally translate words and sentences from [Kenyan] sign language,” she explained.
There is a plethora of innovations that are continuously shared across TikTok. As AfroTech previously told you, @Krysanthum, a mechanical engineer, created a TikTok filter that helps people make straight, middle parts in their hair. Her creation was unsteady at first but after brainstorming, she was able to make it a success.
“I came up with this line thing so that if you turn your head too much to either way you’re going to see green,” she explained. “You don’t wanna see green at all. You just wanna see pink. Then, I created some graphics to help you figure out how to use this filter.”
“Just keep your head straight and go along the line and hopefully this will help us have the most middlest part ever, amen,” she added.
In addition to the middle part filter, she created filters for braid parting. The mission was to support people like her who share the struggle with sectioning their hair during styling it.
Clearly, Black girls in tech stay winning.
