Voice mode broken on Windows v2.1.83 - spacebar types spaces, no audio capture

Status Fixed / completed

Reported on v2.1.83

Maintainer reply ✓ Yes — ashwin-ant

Activity 6 comments · opened Mar 25, 2026 · closed Apr 19, 2026

💡 Likely answer: A maintainer (ashwin-ant, collaborator) responded on this thread — see the highlighted reply below.

Environment

OS: Windows 11 Pro (10.0.26200)
Claude Code version: 2.1.83 (npm install)
Terminal: Windows Terminal + Git Bash
Shell: Git Bash (MINGW64)

Description

Voice mode (/voice) enables successfully and status bar shows "Voice: processing..." but holding spacebar does not record audio — it types spaces into the input instead.

This was also broken on v2.1.81. SoX 14.4.2 is installed (winget install ChrisBagwell.SoX) but does not appear to be used as a fallback on Windows (unlike Linux/macOS).

Steps to reproduce

Launch Claude Code v2.1.83 on Windows 11 via npm
Run /voice to enable voice mode
Status bar shows "Voice: processing..."
Hold spacebar — spaces are typed instead of recording audio
No audio is captured

Expected behavior

Holding spacebar should activate the microphone and record speech for transcription.

Related issues

#31065 — native audio module not bundled for Windows
#30915 — audio-capture.node path hardcoded
#32249 — proposed SoX fallback for Windows
#33504 — voice command fails on Windows CLI
#38960 — space key moves cursor instead of recording

Workaround attempted

Installed SoX via winget install ChrisBagwell.SoX — no change
Tried rebinding push-to-talk key — no change
Downgraded to v2.1.81 — same issue

Notes

The npm-installed version on Windows appears to be missing the audio-capture.node native module entirely, and the SoX fallback path that exists for Linux/macOS is not wired up for Windows.

View original on GitHub ↗

6 Comments

github-actions[bot] · 5 months ago

Found 3 possible duplicate issues:

This issue will be automatically closed as a duplicate in 3 days.

If your issue is a duplicate, please close it and 👍 the existing issue instead
To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

TalkingMonkeyOz · 5 months ago

Update: Version-specific regression confirmed

Voice works correctly on v2.1.81 but is broken on v2.1.83 on the same machine with the same configuration.

Both versions have audio-capture.node bundled at vendor/audio-capture/x64-win32/audio-capture.node (492KB), so this is not a missing native module issue.

The problem appears to be in the spacebar hold-detection / input handling layer — something changed between 2.1.81 and 2.1.83 that causes the spacebar hold to be interpreted as text input (typing spaces) rather than triggering push-to-talk recording.

Environment: Windows Terminal + Git Bash (MINGW64), npm-installed package.

Workaround: Pin to v2.1.81 (npm install -g @anthropic-ai/claude-code@2.1.81)

Likely related to #38960.

jason-ezenwa · 5 months ago

also broken on macOS but works in v2.1.81

konrad-rzonca · 5 months ago

FYI I had the same issue and disabling and reenabling using /voice twice fixed it for me

ashwin-ant collaborator · 4 months ago

This was fixed in v2.1.84 — Fixed voice push-to-talk: holding the voice key no longer leaks characters into the text input. If you're still seeing this in the latest version, please comment with your version and repro and we'll reopen.

github-actions[bot] · 4 months ago

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.