Accessibility & Controller Settings

Probably this will be extended in the future, but SearXR now has:

• a controller-section (see controller-button)
• it will open an accessibility screen with various toggles

For example, offline text-to-speech or offline facetracker (using webcam) can be enabled

Text to Speech

SearXR uses an offline text-to-speech engine ♥
That means it's privacy-friendly, and no audio is sent to any server.
After having tried many offline engines, the eSpeak turned out to be the most convenient.
It's a nice, somewhat-robotic voice, which can be used for multiple (accessibility) future use-cases.

Speech input

Unfortunately after lots of experiments & research, the WebSpeechApi implementation is on hold.
Reasons:

• browsersupport of W3C WebSpeechAPI has stagnated
• the error-rate of 100% javascript speech-recognition-libraries is too high

The above has been described in greater detail in my report

Facetracking, Eyetracking, Handtracking

Below is a small demo of facial navigation using libraries like Webgazer.js and handsfree.js.
It is not meant as a turn-key solution, but as a starting point for specific accessibility needs:

The webcam detects and tracks face-movement/rotations, which get reflected inside SearXR

This allows for interesting accessibility- or exhibition (project on wall) usecases.