[BUG] Voice mode does not recognize Russian speech

Resolved 💬 2 comments Opened Mar 13, 2026 by zkuzma8 Closed Apr 10, 2026

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?

Voice mode (/voice) fails to transcribe Russian speech. Whisper (used for STT) supports Russian, but in practice:

  • Short Russian phrases are either not recognized or transcribed as another language
  • Longer Russian sentences sometimes work, but inconsistently
  • English works fine with the same microphone and environment

Environment:

  • macOS (Darwin 25.2.0)
  • Claude Code (latest)
  • Built-in MacBook microphone

Expected: Russian speech is transcribed correctly, similar to English.

Actual: Russian speech is either ignored, garbled, or misidentified as another language.

What Should Happen?

Russian speech should be transcribed correctly, same as English. Whisper supports Russian - the transcription should recognize the language automatically and produce accurate text regardless of phrase length.

Error Messages/Logs

Steps to Reproduce

  1. Run /voice in Claude Code to enable voice mode
  2. Hold Space and say a short phrase in Russian (e.g. "Покажи статус проекта")
  3. Release Space — the transcription is either empty, garbled, or in a wrong language
  4. Repeat with a longer Russian phrase (2-3 sentences) — result is inconsistent
  5. Switch to English and say "Show project status" — transcription works correctly

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.75 (Claude Code)

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Warp

Additional Information

_No response_

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗