Raspberry Pi Zero Transcribes Muffled Speech in Real-Time with AI

Raspberry Pi
(Image credit: Kevin Lewis)

Maker Kevin Lewis has created a Raspberry Pi-powered solution to understanding speech that’s hard to hear or, in many cases today, muffled by a mask. He’s created a wearable badge that features a small display that generates text from speech using AI to help with accuracy.

According to Lewis, the transcription process is aided using  DeepgramAI an API service which uses AI to transcribe your speech in real-time. The project is flexible in nature and could also serve as a tool for anyone hard of hearing.

Lewis indicates the total cost for the project was around $81 (£60). The badge relies on a Raspberry Pi Zero connected to a small display. A microphone is also necessary for detecting audio while a battery pack makes the unit portable.

See more

The demo video shows a custom interface, controlled via a touchscreen, with a few modes including one that displays only the voice recognized as the wearer and a group mode that displays other voices in various colors. Users can also disable the captioning altogether and display a static message.

To see this Raspberry Pi project in action, check out the original thread shared to Twitter by Kevin Lewis and be sure to follow him for future projects. 

Ash Hill
Freelance News and Features Writer

Ash Hill is a Freelance News and Features Writer at Tom's Hardware US. She manages the Pi projects of the month and much of our daily Raspberry Pi reporting.