Building a Brainstorming Tool with ChatGPT and Whisper
In this video, we explore this concept of leveraging speech to text technologies coupled with LLMs to develop a brainstorming tool. By using ChatGPT and Whisper, the user can talk about an idea, and the discussion will be saved as an MP3 file. Whisper then transcribes the text, and ChatGPT creates an ongoing summary of the thoughts, showing it back to the user. This process can be looped for as long as needed, allowing for the development of ideas through conversation rather than writing.
We walk through the process of building this tool, starting with a Python script to record the audio and then integrating transcription and response functions. The video demonstrates the iterative development of the idea, highlighting the potential of AI in enhancing creativity and thought processes. Whether you're a developer interested in AI or someone looking to expand your brainstorming capabilities, this video offers a look into the future of interactive learning and idea development (at least I think so ;)).
📚 Chapters:
00:00 - Intro
01:23 - Writing the ChatGPT response function
04:30 - Writing the record audio function
07:00 - Writing the transcribe function
09:00 - Creating the brainstorming loop
14:00 - First tests
18:00 - Iterating and improving
28:00 - Final thoughts and discussing the issues with the chat interface
🔗 Links:
Subscribe: / @automatalearninglab
Join Medium: / membership
Tiktok: https://www.tiktok.com/@enkrateialucc...
Twitter: / lucasenkrateia
LinkedIn: / lucas-soares-969044167
Music from www.epidemicsound.com