[BUG] Voice mode shows “listening” on Windows but produces no transcript in any local shell - just prints spaces and doesn't transcribe
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
Claude Code voice mode on Windows enters the “listening” state, but no transcription is produced.
This was working earlier the same day and then stopped suddenly. The only notable change around that time was upgrading PowerShell to 7.6, but the issue also occurs in cmd.exe, so it does not appear to be specific to PowerShell itself.
Environment
OS: Windows 11
Claude Code version: 2.1.81 after reinstall
(previously showed 2.1.83 before reinstall)
Shells tested:
PowerShell 7.6
Command Prompt
Windows Terminal
VS Code integrated terminal
Auth: Claude.ai account login
Microphone permissions: enabled in Windows
Local terminal session: yes
Remote/SSH session: no
Reproduction Steps
Launch Claude Code in a local Windows terminal session
Run /voice
Hold Space to activate push-to-talk
Claude shows “listening”
Speak normally
Release Space
Expected Behaviour
Speech should be transcribed into the prompt input.
Actual Behaviour
Claude shows “listening”, but no transcript appears.
Space is recognised, voice mode appears to activate, but the transcription pipeline seems to produce nothing.
What I Already Tried
Toggled /voice off and on multiple times
Fully closed and reopened Claude Code
Tested in:
Windows Terminal
VS Code integrated terminal
cmd.exe
Confirmed Windows microphone privacy settings are enabled
Confirmed Claude Code had recently accessed the microphone in Windows privacy settings
Tried custom keybindings instead of Space
Reverted keybindings back to default
Reinstalled Claude Code
Same behaviour after reinstall
What Should Happen?
Claude should transcribe my voice input
Error Messages/Logs
NoneSteps to Reproduce
/voice - toggle off
/voice - toggle on
press spacebar - start speaking
Claude Model
Sonnet (default)
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
Claude Code v2.1.83
Platform
Anthropic API
Operating System
Windows
Terminal/Shell
Windows Terminal
Additional Information
_No response_
4 Comments
Found 3 possible duplicate issues:
This issue will be automatically closed as a duplicate in 3 days.
🤖 Generated with Claude Code
Root cause identified and patched — see #38690 for full analysis.
Two bugs in
useVoiceKeybindingHandler:useEffectdependency array includes setState, causing hold-detection counter to reset every renderstopImmediatePropagation()is a no-op in Ink's EventEmitter — space passes through to chat inputWorking patch verified on v2.1.83 (macOS ARM64). Same root cause likely applies to Windows.
This was fixed in v2.1.84 — Fixed voice push-to-talk: holding the voice key no longer leaks characters into the text input, and transcripts now insert at the correct position (regressed in v2.1.83). If you're still seeing this in the latest version, please comment with your version and repro and we'll reopen.
This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.