[FEATURE]: Speech-to-Text input support #31542

Feature hasn't been suggested before.

• I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

Feature Request: Speech-to-Text Input

What do you want to change or add?

I would like to request the addition of a Speech-to-Text (STT) feature directly in the chat input field of opencode.

The idea is to add a microphone button next to the input box. When clicked, it would listen to the user's voice and automatically transcribe the spoken words into text in the input field — just like voice input in modern chat applications.

How it could work

• A microphone icon button is placed inside or beside the input field
• Clicking it activates the browser's native Web Speech API (SpeechRecognition) or an alternative STT service
• The transcribed text is inserted into the input in real time
• The user can then review/edit the text before sending

Benefits

• Faster input: Dictating is often faster than typing, especially for long prompts
• Accessibility: Helps users with physical limitations or those who prefer voice interaction
• Better UX: Brings opencode closer to a natural, hands-free coding assistant experience
• Low overhead: The Web Speech API is built into modern browsers (Chrome, Edge) and requires no additional backend

Suggested Implementation

• Use the browser-native window.SpeechRecognition / window.webkitSpeechRecognition API for a zero-dependency solution
• Optionally support a fallback to an external STT API (e.g., Whisper, Deepgram) via config for better accuracy or offline use
• Add a config option to enable/disable the feature

Why is this beneficial?

Many AI coding assistants are moving toward multimodal input. Adding voice input would make opencode more powerful, accessible, and user-friendly without heavy infrastructure changes.

Thank you for considering this feature!