Live Captions are here for you wherever you are (unless you're not on Google Chrome 😅).
Firstly what are are they? Basically auto generated subtitles for whatever you are watching and they can even translate from one language to another for you in real time.
I love this, a lot of the time I find audio distracting and I also have difficulty hearing; so when I'm watching that latest tech tutorial I use Closed Captions.
You might have noticed most of my articles are inspired by a problem I solved and wanted to share my journey. So an aside, Brave browser a Chromium based browser while having great built in ad blocking features and speed built in actually really lacks accessibility!? (It is literally missing the settings tab "accessibility")
- https://community.brave.com/t/live-caption-from-chrome/250927
- https://techdirectarchive.com/2020/02/09/how-to-use-sysinternals-live-tools/
macOS also has this setting but it can be tricky to get going, it's tied to photobooth for some reason https://support.apple.com/en-gb/guide/mac-help/mchldd11f4fd/mac
There are some other options thankfully, I recently discovered the SpeechRecognisition API via VueUse There is a demo on VueUse it seems fairly accurate.
There are a few more advanced libraries that require a little more setup like openai's whisper; you need to register for an API key. If you're looking for a quick javascript-y way to try it out I've found:
- voice-writing-electron
- and Whisper's own electron project (currently $0) WhisperScript Not sure it's just me but any project with Whisper in the name about code really excites me.
I haven't personally tested but Windows has LiveCaptions too.
I have tried various Linux/Ubuntu solutions with subpar success, I think Chrome is the best bet.
There are some awesome companies making use of this technology to improve you're work life already such as SuperNormal; If you have trouble hearing or don't want to be distracted by taking meeting notes I found this plugin works extremely well. It automatically syncs to your google calendar and creates really beautiful meeting notes with action items and everything. I found the free tier all I needed, this may sound sponsored but it's not. The one caveat check with your legal/IT you are allowed to use any plugins at work, I got bit in the backside by this although it is completely AI driven it may be considered by some 👴 old companies/CTO's as a security breach.