GrahamJenkins
commented
1 month ago Adding my two cents here. With the recently released GPT-4o multimodal capabilities, I can imagine that it would be be interesting to be able to simply chat about the codebase and the problems you are working on while receiving suggestions and/or code help. This would be a great feature to have embedded.
Is your feature request related to a problem? Please describe. I frequently face challenges limitations due to needing to manually type within the chat interface. As someone who is very used to pair programming but works solo more frequently solo I yearn to break down problems out loud but not used to re articulating those same thoughts in the chat. The current lack of a speech-to-text option limits accessibility and ease of use, especially for users who may have disabilities that make typing difficult.
Describe the solution you'd like I would like to propose adding a chat interface option for speech to text. This feature would allow users to speak into their microphone, and cursor would consume the audio as input for the next chat. Ideally, this functionality should be easy to activate, perhaps with a simple toggle switch or voice command to start dictation. The interface should also include feedback mechanisms, such as visual cues or gentle sound notifications, to confirm that speech is being recorded and converted correctly.
Additional context Add any other context or screenshots about the feature request here.