Use Cases › Documentation with Voice in OpenCode
📝
Documentation with Voice in OpenCode
The Problem
Writing documentation is often neglected because it feels separate from coding. Stopping to type docs breaks the development flow.
The Solution
Voice lets you document as you go. Describe functions and APIs verbally with /ptt, have the AI generate formatted documentation, and review it via TTS.
Workflow
- After implementing a function, type /ptt instead of reaching for the keyboard.
- Speak the documentation: "This function takes a user ID and returns their order history, sorted by date descending. It handles pagination internally."
- The AI formats this into proper JSDoc or Python docstring format.
- Review the generated docs via TTS: /tts-speak the doc output.
- Interrupt to refine: "Add an example of the return value format."
- The AI updates the documentation based on your spoken feedback.
Tips
• Voice documentation is faster than typing for most developers — speaking is 3x faster than typing on average.
• Whisper handles technical descriptions well. For API-specific terms, set whisperModel to "small" minimum.
• The AI is good at translating conversational descriptions into formal documentation standards (JSDoc, Sphinx, etc.).