Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Important Feature Request(s) #131

Closed
vaibhavard opened this issue Jun 22, 2023 · 6 comments
Closed

Important Feature Request(s) #131

vaibhavard opened this issue Jun 22, 2023 · 6 comments

Comments

@vaibhavard
Copy link

vaibhavard commented Jun 22, 2023

1.One click Conversation Clear (Since the history - conversation - clear button does not work for me
image )

2.Bing AI (Can replace and enhance "/react" , Bard Integration
3.Increased ASR Listening Duration (So you dont have to click microphone button again and again)
4.OpenAI Embedding Based PDF Q&A

These are from the bugs i encountered when i was using the app...

@enricoros
Copy link
Owner

enricoros commented Jun 23, 2023

One click Conversation Clear

Updated the History menu to be more clear - it's the Sent Messages history. To clear the chat, click on the "Reset" button in the menu:
image

@enricoros
Copy link
Owner

enricoros commented Jun 23, 2023

OpenAI Embedding Based PDF Q&A
There are 2PRs for this, see video of #117, and #122. Give it a look. It's hard to productize without impacting existing users.

Bing AI (Can replace and enhance "/react" , Bard Integration
I don't know if Bing AI has an API. Bard as a model can be integrated.

Increased ASR Listening Duration
Unfortunately the Chrome API closes the connection so quickly as soon as silence is detected. Hint, use CTRL+M to avoid clicking again. Otherwise we'll need to re-arm the Mic using our own timers. Not easy to get it right.

@vaibhavard
Copy link
Author

Thanks for the quick response 😀. This project is really very helpful!

I don't know if Bing AI has an API.

You can use this : https://github.com/waylaidwanderer/node-chatgpt-api ,https://github.com/tuhinpal/bingchat-api

Unfortunately the Chrome API closes the connection so quickly as soon as silence is detected. Hint, use CTRL+M to avoid clicking again. Otherwise we'll need to re-arm the Mic using our own timers. Not easy to get it right.

Cant you restart it on our own backend?..Maybe, you can experiment with big-agi\src\common\components\useSpeechRecognition.ts. I tried the togglerecording but it does not work. (PS:I am very new to TYPESCRIPT)

        instance.onresult = (event) => {
          if (!event?.results?.length) return;
          let transcript = event.results[event.results.length - 1][0].transcript;
          // shall we have these smarts?
          transcript = (transcript || '')
            .replaceAll(' comma', ',')
            .replaceAll(' exclamation mark', '!')
            .replaceAll(' period', '.')
            .replaceAll(' question mark', '?');
          if (transcript)
            onResultCallback(transcript);
          toggleRecording(); // restarts recording (Currently does not work)

        };

@enricoros
Copy link
Owner

enricoros commented Jun 23, 2023

I'm also a big user of the Mic, have a couple of ideas in mind.
One is to set the Continuous operation to true (may not be now), which should listen for more than one sentence.
I may even do a UI fix to show the partial sentence being written as one speak (as an overlay, to avoid expensive repaints to the base component)

@enricoros
Copy link
Owner

3.Increased ASR Listening Duration (So you dont have to click microphone button again and again)

@vaibhavard Mic support has been overhauled. Works best on Chrome/Windows. LMK if it works for you.

@vaibhavard
Copy link
Author

Thanks A lot for this feature ..Sorry for the delay in reply...The feature works as intended😀😀

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants