OpenAI Whisper Hackathon summary: the participant’s story

We asked our Open AI Whisper Hackathon’s winners to share with us their journey and experience during the Transform with Transformers Hackathon, which took place on the 28th of October 2022.

It was an impressive event, with over 1600 AI engineers from all around the world, who put all of their hands on board to build and create revolutionary AI apps using the newest speech-to-text technology - Whisper by OpenAI.

OpenAI's Whisper is a voice recognition technology that, since it came out, instantly became a game changer. The model has been trained for over 680,000 hours (about 77 years! ), making it robust and accurate, approaching human-level accuracy.

So… without any further ado, read what winners of the Whisper Open AI hackathon, organized, of course, by LabLab.ai, want to share!

🥇 1st place: Nikhil Sehgal, Team Blue

I joined the OpenAI Whisper hackathon after seeing a post on LinkedIn. It was my first time working in a team of 6 people worldwide, which was pretty daunting. However, we all had the same passion and a mission of achieving a common objective.

The process was surprisingly seamless and unbelievably satisfying. I was impressed with how the whole Lablab.ai team orchestrated the Hackathon.

Our prototype was built to solve a problem for many people now - investing is complicated and developing a good strategy is time-consuming. We used OpenAI's Whisper, GPT-3 & Codex to build a solution that listens to a YouTube clip about a trading strategy and generates a trading algorithm based on it. That means this person can now test & implement more strategies, thereby increasing their likelihood of finding a plan that works for them. They ultimately achieve financial freedom!

🥉 3rd place: Georgy Struchkov, Team Boss

I found out about Lablab through an advertisement for a Whisper AI Hackathon on Instagram. I signed up immediately because I've always been interested in participating in such an event. I waited to meet my team right away, though, because the first team of randoms fell apart shortly before the start of the Hackathon. We found teammates just before the beginning and quickly found a common language. It was fascinating to work with guys from all over the world— it significantly increased the level of my soft skills.

The Lablab.ai team gave us an excellent opportunity to prove ourselves and do a cool project. They were always ready to help and answer questions about the project. Whisper technology is the most advanced Speech-to-text model to date. The Lablab team enabled us to level out people’s stuttering and direct pure speech recognition. The Lablab infrastructure also greatly simplified the work— we connected to a server with an API model, transferring the need for deployment from our local machines to the Lablab.ai servers.

In the end, the work was a success! We did an exciting project that can help people with stuttering quickly and conveniently translate their speech into text without much effort.

😊 Extra story of participant: Omar, Team Chasers (winner of the Wolves Summit Hackathon)

My story with Lablab.ai began when I was scrolling down on LinkedIn and saw a post about the Wolves Summit Hackathon. I gathered a team of my students based on their technical knowledge and communication skills, and we joined the AI Hackathon. It wasn't easy, but we worked so hard that we got 1st place on the Hackathon with the "Chase the Fire" project. For this project, we've used OpenAI's Whisper and a Python library called "Sikit Learning."

It works on predicting forest fires and their directions to lure the animals away and give us time to keep our ecosystem safe.

At the OpenAI Whisper Hackathon, we used the whisper technology to do a project that can hear the user talking in any language except English and then translate it into English. Also, we spoke about the speaker, which can help many people. Also, we uploaded code on a Raspberry Pi, that can be used as a device with a microphone and speakers. Then when we found the community of LabLab.ai so helpful and supportive, this gave us the courage to join the next Hackathon!

✍️ Final words

We would like to thank all the participants of the OpenAI Whisper Hackathon for their time, energy, and creativity! Your work shows people, that it’s not impossible to start their journey with AI technology at any point of your career and you can make a tool that has a real impact on the world!

We are happy that we could bring together many talented people worldwide and give them the opportunity to show their spectacular skills. We are looking forward to seeing you at our next Hackathon!