TTS readback of responses + voice mode for Remote Control sessions
Preflight Checklist
- [x] I have searched existing requests and this feature hasn't been requested yet
- [x] This is a single feature request (not multiple features)
Problem Statement
When using Claude Code remotely from a phone via Remote Control, there is no voice capability — all interaction requires
typing on a small screen. Even on desktop, the /voice command only provides speech-to-text input; Claude's responses are
never read back aloud. This makes hands-free or eyes-free coding workflows impossible. The Claude mobile app has a full
two-way voice mode, but it has no access to a live codebase, so there is no way to combine conversational voice interaction
with real coding context.
Proposed Solution
- Add TTS readback of Claude's responses when /voice mode is active on desktop — after Claude finishes a response, it
should be read aloud automatically (with an option to interrupt or skip).
- Expose voice input and TTS readback in the Remote Control web interface, so users connecting from a phone or tablet can
have the same experience.
Alternative Solutions
- Piping Claude Code output to an external TTS engine (say on macOS, espeak/piper on Linux) — possible but not integrated,
requires custom scripting, and breaks the native UX.
- Claude mobile voice mode + Project file uploads — has full two-way voice but no live codebase access; context goes stale
immediately.
- No satisfactory end-to-end solution exists today.
Priority
Low - Nice to have
Feature Category
Developer tools/SDK
Use Case Example
- I'm away from my desk and connect to my running Claude Code session via Remote Control on my phone.
- I want to ask Claude to explain a module or walk through a bug.
- With this feature, I could speak my question, hear the response read back, and have a back-and-forth conversation about
the code — completely hands-free.
- This would make code review, exploration, and planning practical while away from a keyboard.
Additional Context
The Claude mobile app already demonstrates that two-way voice works well for general conversation. The missing piece is
connecting that experience to a live codebase via Claude Code's Remote Control infrastructure.
This issue has 8 comments on GitHub. Read the full discussion on GitHub ↗