Inspiration
I was inspired when I saw a social media post contemplating the lack of alt text available for many images across countless websites. Even on social media, many people are pushing for regular users to attach "alt text" to any media they upload. However, many people fail to provide accurate/significant alt text, if at all.
What it does
This extension communicates with a vision compute (image "describer") AI through a third-party API. This API responds with a detailed description of the image. This description is then passed to a speech synthesis (text-to-speech, aka "TTS") API, which responds with the audio explaining the image.
How I built it
This extension was made in Javascript, with some aspects like the popup/settings page being made with HTML and CSS.
Challenges I ran into
Manifest V3 is Google Chrome's new "extension platform", and it is notoriously terrible. This new manifest version may bring important security changes to the platform, but it makes creating certain extensions unbelievably difficult and tedious. I could have easily made this extension in the deprecated "V2" manifest version, but such an extension would not be supported on the Chrome store.
Accomplishments that I'm proud of
I'm proud to present a (mostly) working accessibility extension that will soon be published on the Google Chrome Extensions store/catalog. I hope the extension will make browsing the web an easier and more comfortable experience for those with visual impairments.
What I learned
- Learning some basic HTML
- Learning some (very) basic CSS
- Learning some JS
What's next for AltSpeakify
- Bug fixes! So many bug fixes!
- Better "voice ID" picker for ElevenLabs (drop-down menu instead of text box)
- Configurable volume settings for the audio (error sound, wait sound, describe TTS/alt txt, etc.)
Built With
- 11labs
- astica
- chrome
- chromium
- css
- elevenlabs
- extension
- html
- javascript
- tts

Log in or sign up for Devpost to join the conversation.