andygmassey

andygmassey / telephone-and-conversation-transcriber

Public

Real-time speech-to-text caption appliance for a deaf user. Raspberry Pi + 10" touchscreen that transcribes phone calls and room conversation in near real-time.

accessibility assistive-technology deaf deepgram pyqt6
90
6
100% credibility
Found Feb 11, 2026 at 40 stars 2x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A Raspberry Pi touchscreen application that transcribes phone calls and room conversations into real-time captions for hearing-impaired users.

How It Works

1
👂 Discover the Helper

You hear about this friendly tool that turns spoken words into readable text on a screen to help a loved one who is hard of hearing follow phone calls and chats.

2
🛒 Gather Simple Parts

Pick up a small touchscreen computer, a room microphone, a phone audio adapter, and download the easy program.

3
💻 Prepare the Setup

Put the program on the small computer following the friendly guide, like setting up a new gadget.

4
🔌 Connect Everything

Plug the screen, microphone, and phone adapter into the computer – it all fits neatly together.

5
✨ Power On Magic

Turn it on next to the phone, and instantly see words from conversations appear live in big, clear letters on the screen.

6
👆 Make It Perfect

Tap the screen to pick bigger text, fun colors, or switch listening modes if needed – super simple.

😊 Clear Conversations

Now your loved one smiles, reading every word during calls and talks without missing a thing – it just works reliably.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 40 to 90 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is telephone-and-conversation-transcriber?

This Python project turns a Raspberry Pi 5 with a 10-inch touchscreen into a real-time speech-to-text captioner for deaf users, transcribing landline phone calls via USB recorder and room conversations via conference mic. It displays large, scrollable captions with touch controls for font size, colors, and online/offline modes, falling back to local engines like Vosk or faster-whisper when cloud services like Deepgram are unavailable. Idle time shows a flip-clock that dims at night—plug it in next to the phone, and it runs headless via systemd.

Why is it gaining traction?

It stands out with automatic phone call detection that switches audio sources seamlessly, bulletproof watchdogs for restarts on crashes or network drops, and hybrid online/offline real-time speech-to-text that prioritizes low latency (200ms cloud, 300ms local). Developers dig the ready-to-deploy setup with clear hardware lists and quick-start scripts, making it a solid base for custom real-time transcription projects on GitHub without reinventing audio handling or UI.

Who should use this?

Raspberry Pi enthusiasts building accessibility aids for elderly relatives, hardware hackers prototyping real-time speech-to-text devices for Linux, or indie devs needing an offline-capable captioner for conference rooms or calls. It's ideal if you're tired of flaky cloud-only APIs and want touch-friendly, reliable local transcription out of the box.

Verdict

Grab it if you're in the niche—strong docs and MIT license make forking easy, despite 14 stars and 1.0% credibility signaling early maturity. Test on your Pi setup; it's a functional prototype worth iterating on for production real-time speech-to-text needs.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.