I talk a lot, mostly with myself. For the last few years, I’ve made a lot of voice memos. I find it an ideal way to organize my thoughts while taking a short walking break or commuting home.

One big downside is that voice memos are not as easily searchable. So, searching back for specific information is challenging.

In this talk, I explain the Whisper model from OpenAI and create a system with Azure OpenAI to get transcripts from my voice memos. I also use the GPT4 model to enhance the transcript with extra information.

When you leave the session, you understand what Whisper and GTP4 can do for you and hopefully go inspired to build your own workflow.