Why I run speech-to-text locally instead of calling a cloud API

Why I run speech-to-text locally instead of calling a cloud API Yesterday I wrote about deploying gemma, bge-m3, and whisper on a single server without enough VRAM for all three. This post is about wh...