Prettify the UI

[theoryspace]

• Use toggle switch for switching between GPT4 and Gemini
• Add user prompt and result transcription log box
• Add ellipses and retry if rate limited (code 429 too many requests)
• Restart dictation if error comes back with silence is too long
• Move inline styles to css file
• Catch errors in gemini.ts and log to console

[theoryspace]

Thanks @gregsadetsky for this wonderful demo. Here's my little contribution to make the UI a little nicer and more intuitive to use.

[gregsadetsky]

really great work, thanks a lot!!

just tried & reviewed the code, looks like a fantastic improvement

if you're up for it, it would be really great if the API keys (for either/both openai & gemini) could be input directly on the page i.e. two <input type="text"> somewhere ...? at the top of the page?

that way I could push this project live (i.e. deploy it to a url) and anyone with API keys could try it -- not just folks who are able to install it. let me know if that makes sense and if you'd like to tackle that! no pressure of course.

the api keys directly on the page are just the next steps I had in my todo :-) thanks again!!