B2 Bachelor VoiSum Use AI to summarize voice messages

In today’s fast-paced world, we’re aiming to transform how people consume voice messages by creating an AI-powered solution that cuts through the noise. Our tool takes those rambling voice notes and delivers just the essential content - no fluff, no filler words, just the key points you need. By combining speech-to-text, smart summarization, and voice synthesis, we’re not just saving time - we’re making voice communication more efficient while keeping it personal.

So, lets break down what our software can actually do? First up, we’ve got this cool transcribe-and-summarize combo - just hit the endpoint and boom! You get both a full transcript of your audio AND a neat summary. No more dealing with them separately!

Want just the highlights? Our keywords feature has your back, pulling just the bullet points out of your message. And for those times when you need specific control, we built this sweet custom-length summary feature - you decide how long or short you want it.

Testing it out is super easy too! Record something (like that random story your friend told you), and watch our API work its magic. Or if you’ve already got some audio files collecting dust on your device, just upload them and let our features do their thing! It’s basically like having a personal assistant who’s really good at cutting through the fluff - except it’s all AI-powered and way cooler!

Our journey kicked off when Prof. Tobias Lenz dropped this awesome idea in our laps - talk about starting with a bang! First things first, we got organized with a GitHub repo and set up a Kanban board to keep our workflow smooth (because who doesn’t love checking off those task boxes, right?). Before diving in headfirst, we went full detective mode on existing AI models - to make sure we weren’t reinventing the wheel. Once we had the lay of the land, we sketched out our game plan and started refining it. The tech stack hunt was real - we had to find the perfect combo that would make our project work.

We kept things agile and flexible with weekly catchups (complete with endless coffee runs). It was all about that “build, test, improve” life - tweaking and fine-tuning until we had something that wasn’t just functional, but actually made sense to use. No one likes a clunky app, after all! This approach meant we could pivot faster than a TikTok trend whenever we got feedback, keeping our project fresh and on point.

We divided our Team into 2 specialized groups, one to build the API, the backbone of our software and the other to explore and build different Frontend solutions.