UNIT.City — місце, де люди працюють... КРАЩЕ! Обирай свій простір просто зараз 👉
Ігор Вишневський AI Eng
2 February 2025, 10:01
2025-02-02
A schoolboy from Pokrovsk has developed an AI dictionary for learning sign language. How the invention works
A student of Pokrovskaya School No. 9, tenth grader Mykyta Telichko, who is currently in the United Kingdom, developed a computer AI dictionary for learning sign language, winning a gold medal at the Small Academy of Sciences' innovative projects competition.
A student of Pokrovskaya School No. 9, tenth grader Mykyta Telichko, who is currently in the United Kingdom, developed a computer AI dictionary for learning sign language, winning a gold medal at the Small Academy of Sciences' innovative projects competition.
The development helps people with hearing impairments understand each other and quickly master sign language, writes Suspilne.
According to Mykyta Telichko, he was inspired to develop it by his brother, who uses a hearing aid.
«My younger brother Makar, who is 12 years old, has special needs — he can’t hear very high frequencies and therefore uses hearing aids. I thought: how can I help people like Makar? I analyzed and realized that it is very difficult to learn sign languages,» says Nikita.
Last year, thanks to the invention of an interactive dictionary using computer vision and open databases, he won first place in the «Sustainable Living Space» nomination in the final stage of the All-Ukrainian competition of youth scientific and technical projects «InventorUA».
According to Mykyta, he began to be interested in robotics and IT technologies at the age of 7-8.
«I wanted to learn sign language, for example, whether it is necessary — what are the current solutions. You just go to the site, write a word, find a video — how to show this word. And you try to reproduce it, but you don’t get any feedback. That is, you just try to repeat, but you don’t understand whether you are showing correctly. My project helps to check whether they can understand you, whether you are showing correctly. And it’s also faster to learn it, because when you show it yourself, you remember it better,» explains Mykyta Telichko.
The key components of the solution in his project are the use of Python and MediaPipe for gesture recognition, the integration of the WLASL database, and the development of an intuitive user interface.
The prototype of the student’s dictionary has a gesture recognition system via a webcam, a comparison function with reference samples, and a visual display of the similarity of the performed gesture with the reference one, says Daria Lytvyn, a computer science teacher at Pokrovskaya School No. 9 and Mykyta’s scientific supervisor.
«Mykyta did it in such a way that he used artificial intelligence to recognize movements — the camera reads the movements, compares them with the original image, and displays the percentage of coincidence on the screen. 100% is a perfect match, 90-95% is already well done. If it shows 20-30%, you need to repeat to improve your level,» said Darya Lytvyn.
The young man’s invention was tested by the Association of the Deaf in one of the cities of Great Britain. Based on their feedback, Mykyta says, he thought about adding more languages to his development.
«They said the project is interesting. I want to add more languages, including British. American and British sign languages are very different, that is, radically different,» notes Mykyta.
According to Darya Lytvyn, Mykyta Telichka was invited to the national team to represent Ukraine at international competitions.
«This is already a startup, there is still work to be done, but a large amount of work has been done and he has already planned what he will improve. As soon as some international exhibition or competition appears and Mykyta’s project falls into this category, the guy will participate. He is currently preparing all the specifications and technical documentation in English, he speaks it very well,» the teacher said.
At the same time, inventions are a thing for all ages. dev.ua wrote that an 80-year-old inventor from Dnipro created a bicycle generator that can generate 5 hours of electricity in 15 minutes.
A schoolboy from Kyiv region came up with a way to protect drones from enemy electronic warfare. Plans are underway to launch the development into mass production
Як нейромережі бачать вільну та незалежну Україну? Тест dev.ua
Нейронні мережі для генерації зображень бачать світ по-своєму, їхню логіку зрозуміти часом зовсім неможливо. Але таки хочеться. На честь Дня Незалежності України редакція dev.ua вирішила провести невеликий експеримент.
Ми задали чотирьом різним нейронним мережам п’ять однакових запитів: «прапор України», «День Незалежності України», «український Крим», «перемога України» та «українці». Отриманими результатами ми ділимося з вами нижче.
У TikTok тепер можна генерувати фон за допомогою нейромережі. Ми протестували її та ділимося результатами
У TikTok з’явилася нова функція «Розумний фон». З її допомогою як фон для тіктоків можна підставляти згенеровані нейромережею зображення. Редакція dev.ua протестувала цю технологію і ділиться своїми враженнями.