Invention Title:

SPEECH-TO-TEXT CAPTIONING SYSTEM

Publication number:

US20250149042

Publication date:
Section:

Physics

Class:

G10L15/26

Inventor:

Applicant:

Drawings (4 of 6)

Smart overview of the Invention

An innovative system for real-time speech-to-text captioning is introduced, combining extended reality (XR) eyewear with a computational eyewear case. Designed to assist individuals with hearing challenges, this system provides captions in the user's field of view by converting spoken words into text. The eyewear component includes microphones, a sensor, a display system, a wireless transceiver, and a processor to capture and process audio. The eyewear case supports these functions with additional processing power and wireless communication capabilities.

Background

Existing hearing aid technologies often struggle with speech comprehension in noisy or multi-speaker environments. While they amplify sound, they do not necessarily improve understanding for those with severe hearing loss. Captioning glasses have emerged as a solution by providing visual text cues during conversations. However, achieving accurate real-time transcription requires robust processing and communication systems, which are often hindered by the limitations of smartphones used in current XR products.

Innovative Solution

This system addresses these challenges by utilizing a dedicated computational eyewear case rather than relying on smartphones. The case accommodates hardware for running real-time speech-to-text systems and ensures reliable connectivity through dedicated wireless antennas. It also houses a large-capacity battery to support the processing demands. By offloading tasks from the eyewear to the case, the design becomes simpler and more cost-effective, resulting in lighter and more comfortable glasses.

Technical Features

The XR eyewear features a multi-sensor microphone array that enhances the signal-to-noise ratio, improving transcription accuracy. A smart sensor detects when the wearer is speaking to prevent unnecessary captioning of their own voice. The eyewear case functions as both a remote microphone and an accessory for assistive hearing devices. These features collectively enhance user experience by providing accurate captions and maintaining battery efficiency.

User Experience

The system emphasizes simplicity and reliability, with pre-paired components that allow users to begin using the glasses immediately upon wearing them. By leveraging sight to enhance speech comprehension, this user-friendly solution offers an effective assistive technology for those with hearing challenges. The design facilitates broader acceptance and adoption due to its ease of use and aesthetic appeal.