[Feature Request]: Voice-to-Text Input in Desktop App

[emranpy]

What problem are you trying to solve?

Typing out long, complex prompts or code instructions into the Desktop app can be slow and tedious, especially for detailed tasks. Having to manually type everything limits accessibility for users who prefer dictation or need hands free interaction while working.

What would you like to happen?

I would like a voice-to-text (Speech-to-Text) input feature added to the Desktop app UI:

1. Microphone UI Button: Add an interactive microphone icon button next to the main prompt input field.
2. Local Audio Capture: Clicking the microphone button records audio using standard Web Audio APIs (getUserMedia).
3. Local Whisper Integration: Convert the audio to text locally using a lightweight open-source Speech-to-Text model like whisper.cpp (using the ~145MB Base model) to keep the app completely free, private, and offline-capable.
4. Text Insertion: Automatically insert the transcribed text into the prompt text box so the user can review and edit before sending.

Area

Desktop app

Contribution

• I would be willing to open a PR for this