Voice Cockpit: On-Device Voice Commands for Radiology Apps

Voice Cockpit: On-Device Voice Commands for Radiology AppsVoice Cockpit: On-Device Voice Commands for Radiology Apps

The network for creativity

Join 1.25M professional creatives like you

Connect with clients, get discovered, and run your business 100% commission-free

Creatives on Contra have earned over $150M and we are just getting started

Back to feedPost

Irdi Zeneli

• Oct 15

Voice Cockpit: hands-free voice commands for a radiology web app. On-device speech recognition (sherpa-onnx in WebAssembly), commands fire on partial transcripts, and a confirm step guards destructive actions. 0% false triggers while dictating, 82% command recall on 128 test utterances. TypeScript + Vite. github.com/IrdiZ/cutting-edge-voice-agent

Drew Sepeczi

• 1d

Vault Zero — AI voice agents for service businesses.

Completed work

AI Automation AI Voice Development Full Stack Development Next.js Twilio Vapi

Dmitriy Chugunov

• 1d

Everyone on your team sends 5-minute voice notes. Nobody has time to re-listen to them.

So I built a fix — and it's now live on Contra:

🎙️ Telegram Transcriber — an n8n workflow template that: • transcribes voice messages, audio and video right in Telegram (2–5 sec) • auto-detects the spoken language and delivers the transcript in the language YOU pick — 90+ languages, per-line AI translation, timestamps preserved • exports Markdown, TXT or .docx • runs 100% on YOUR n8n + YOUR Telegram bot — no subscriptions, no third-party SaaS touching your data

55 nodes, battle-tested: size guards, double-tap protection, auto-retries, operator alerts. Setup takes 10–15 minutes with zero coding — full step-by-step guide included.

Runs at $0/month on Groq's free tier.

👉 https://contra.com/products/DDYfrwTp-telegram-transcriber-bot-n8n-template-whisperai-translation

Questions? Message me — happy to tailor it for your workflow.