I literally cannot take notes in class, so I built an AI to do it for me.
Using OpenAI's Whisper API and Google Gemini, our application troplaflemme turns audio files of lectures into summarized notes.
GitHub for troplaflemme (make sure to star it):
https://github.com/KhanradCoder/troplaflemme
OpenAI's Whisper:
https://github.com/openai/whisper
Google Gemini Quickstart:
https://ai.google.dev/tutorials/python_quickstart
This video is a walkthrough of me building the basis of troplaflemme, a flask app that takes audio files, transcribes them using Whisper, and then summarizes the transcript via Google Gemini prompt engineering. If you have any questions or feature suggestions, let me know down below.
0:00 Intro
1:09 Audio transcription with OpenAI's Whisper
3:19 Literally just me yapping about the FitnessGram Pacer Test
4:20 Testing audio transcription
5:45 Why use Gemini/LLM?
7:09 Google Gemini prompt engineering
7:57 Testing Gemini's summarization
9:16 Combining the transcription and summary
11:22 Final test
12:45 Conclusion
Get VARO Money
https://varomoney.com/r/?r=Adam3090
Follow Me:
Twitter: https://twitter.com/theAdamEubanks
Instagram: https://instagram.com/theadameubanks/
GitHub: https://github.com/KhanradCoder
21 Comments