How to transcribe audio privately, without uploading it
Short answer
Speech-to-text models such as Whisper can run inside a modern browser using WebGPU or WebAssembly, so the audio never leaves your computer. Load the model once, drop in a recording and wait; larger models are more accurate but slower.
- Open the transcriber in a current Chrome, Edge or Safari.
- Choose a model size (smaller = faster, larger = more accurate).
- Drop in the audio file; it is processed on your device.
- Export the text or subtitles.
Hearlark Pro: Meeting notes from recordings, on your laptop
from $10/monthHearlark makes meeting notes from recordings on your laptop: free up to 30 minutes; Hearlark Pro from $10/month for longer recordings.
Secure checkout · 14-day fix-or-refund if it doesn't work as described (terms)Questions
Does it work offline?
After the model has downloaded once, transcription runs locally.
Free: the Free API Starter Kit
Copy-paste calls for 4 APIs with free daily tiers (CVE priority, VIN recalls, US duty rates, email checks), then 5 short emails on KEV/EPSS, MCP and free data.