Take a look at the webpage I've further modified based on your design.

BadgerHobbs / Dave

DAVE - Your Digital Assistant with Voice Empowerment

https://dave.badgerhobbs.com

GNU General Public License v3.0

2 stars 0 forks source link

Take a look at the webpage I've further modified based on your design. #2

Open RickyCong opened 11 months ago

RickyCong commented 11 months ago

Take a look at the webpage I've further modified based on your design. https://116.62.162.233/?apiKey=sk-LSswfHA2ba3h00OV2a6388A5D6Fa4f87Aa488bA529Ea562b

BadgerHobbs commented 11 months ago

You may want to remove and revoke your API key in the URL as you've just made it public.

RickyCong commented 11 months ago

It's okay, I've set a timer for this key. It's not an official API, but rather an API that has been repackaged by a Chinese entity. I'm not a front-end engineer; all the code was modified based on chtgpt.

RickyCong commented 11 months ago

https://www.gptapi.us/

BadgerHobbs commented 11 months ago

https://www.gptapi.us/

Ah I understand, it looks like a reseller/proxy service for OpenAI APIs.

RickyCong commented 11 months ago

exactly

BadgerHobbs commented 11 months ago

Just tried a little demo, works nicely.

My suggestion would be to perhaps make it say something in the chat box at the start or have it be invisible when no messages. Before I asked a question I was wondering what the white box was for.

RickyCong commented 11 months ago

https://api.nextweb.fun/

Here's another reseller/proxy service provider

RickyCong commented 11 months ago

Yes, I don’t know how to write web page code; I am a binary reverse engineering engineer. The function I wanted to implement was for practicing spoken language. However, if there is no dialogue box, I won't know if it can correctly recognize what I said and what it said. You are a professional when it comes to web development. If you can, you might be able to perfect this feature.

BadgerHobbs commented 11 months ago

Yes, I don’t know how to write web page code; I am a binary reverse engineering engineer. The function I wanted to implement was for practicing spoken language. However, if there is no dialogue box, I won't know if it can correctly recognize what I said and what it said. You are a professional when it comes to web development. If you can, you might be able to perfect this feature.

If you just feed the html to ChatGPT with my suggestion, that would probably implement it for you. Worth giving it a go and learning/seeing what it changes to make it work.

RickyCong commented 11 months ago

I'm using it to practice speaking, and most scenarios will definitely be on mobile devices. I've only made a version that's compatible with my device; I've done my best.

001 002

BadgerHobbs commented 11 months ago

The code is mobile responsive, so works on all devices without issue, both desktop and smaller phones.

RickyCong commented 11 months ago

But it seems there is a bug in this code; it can't successfully record audio on Safari for Apple devices.

BadgerHobbs commented 11 months ago

But it seems there is a bug in this code; it can't successfully record audio on Safari for Apple devices.

Interesting. That may be a bug with my code implementation. Feel free to create a bug issue for it.

RickyCong commented 11 months ago

Yes, you could try to reproduce this bug.

But it seems there is a bug in this code; it can't successfully record audio on Safari for Apple devices.

Interesting. That may be a bug with my code implementation. Feel free to create a bug issue for it.