I've been working on a web service to provide speech-to-text transcriptions to Chief of Staff.

Originally, I presumed some GPU-muscled serverless function would be the way to go. However, on the services I looked at the time the model loading meant extended delays in getting shorter transcriptions done.

I wondered: "How cheaply can a persistant instance of Whisper as a webservice be had, and how fast could it be to deliver reasonable quality?"

Ahmet Oner has a great starting place with his `whispe…

Hello and welcome to the new Chief of Staff blog.

The goal for this blog is to share updates to the product, provide information about some of the technical work done to create it and share interesting related content and media.