Skip to content

Experiment with adding OpenAI Realtime WebRTC PTT button to popup UI #18

@paulpv

Description

@paulpv

This seems like a bit of a stretch, but this whole project/repo is a side-effect of https://github.com/LookAtWhatAiCanDo/Codeoba.

Codeoba is a Kotlin Multi Platform implementation.

It would be harder to implement v1 Copilot Task Session DOM text parsing in Kotlin.
This project/repo is a convenient way to both more rapidly prototype both behavior and implementation of a Copilot TTS.

But it occurred to me while building this that maybe it might be interesting to try to put a PushToTalk button in this extension UI similar to one in Codeoba.

Concerns:

  1. I have not even gotten Codeoba's MCP working yet (there are some issues with OpenAI's MCP Client implementation), so I am not sure how productive that will be. If PTT MCP Client -> GitHub MCP Server is not working there then it definitely won't work here.
  2. On a Mac (and I am sure there is something similar on Windows) a user can easily enter SpeechToText mode by double pressing the Globe/Fn key. The existence of that essentially makes a PTT button pretty pointless.

There would need to be some usable and useful reason to put an OpenAI Realtime WebRTC PTT button in this extension's popup UI, and I am not sure if I have come up with that yet. :/

But, it might still be cool to experiment/prototype/play with.

Thus, the "Experiment" in the title of this Issue.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions