I add working voice input to your existing web app: push-to-talk dictation, hands-free voice commands, or both. I built a voice-driven radiology reporting product that a radiology team uses daily, and Voice Cockpit, which runs speech recognition fully in the browser with 0% false triggers while dictating.

What you get: voice input on the screens you choose, the commands you need with a confirm step for risky actions, tests so dictation never fires a command by accident, and a pull request with a short handover note.

I pick the speech engine to fit you: the browser's speech API for speed, a cloud provider for accuracy, or on-device models when audio must not leave the user's machine (health, legal). Works with React, Next.js, Vue or plain TypeScript. EU-based and GDPR-aware.