I made a free AI audio transcriber using the Google Gemini API! I was going to post about this a while back, but I forgot. I launched it on Peerlist today, which reminded me! https://peerlist.io/bestcodes/project/free-audio-transcriber
The code is open source on GitHub and deployed (of course) on Vercel: https://free-audio-transcription.vercel.app/
It uses Next.js + Google Generative AI. I might switch to the Vercel AI SDK soon 
Try it out and let me know what you think!
Thanks for sharing this! I can't wait to try it out soon 
Also, curious about the transition to AI SDK soon 
I feel like the AI SDK will integrate better with React and Next.js, especially for streaming. Not sure if I'll switch yet, since things seem to be working fine right now, but I will probably do some tests with it 
Let us know if you have any specific questions 
I'm using the ai SDK now with @ai-sdk/google on a side-project and I love it! It makes streaming and tool calling so much easier, especially with all the things it does for you in React and Next.js.
When I finish up all my side-projects, I will be switching my audio transcriber to the AI SDK for sure 
I updated the repo with a README for developers! Now devs can see how it works. There is a flowchart to show the logic, making it easy to understand.
Check it out!
This is awesome. Thanks, !
