Speech model sizes explained: tiny to turbo, and which one your browser can run
8 min Whisper model sizes from tiny to large and turbo: what parameters mean, the VRAM and speed figures from OpenAI's own table, and what each costs in a browser.
Hardware.
WebGPU, graphics cards and what your laptop can actually run.
What is WebGPU, and why your browser can now run a speech model
8 min WebGPU lets web pages use your graphics card for maths, not just drawing. Which browsers have it, what can block it, and why it matters for speech recognition.