Speech-to-Text Linux FAQs

Question 1

How does it integrate with Claude Code?

Accepted Answer

It integrates as an MCP server. Claude Code needs to run within a Tmux session, allowing the tool to directly inject transcribed text into Claude's input. You activate voice input with a Right Ctrl key press.

Question 2

What technology does it use for speech transcription?

Accepted Answer

By default, Speech-to-Text Linux utilizes the Whisper tiny model for on-device speech-to-text transcription. This ensures local processing, privacy, and efficient performance without relying on cloud services.

Question 3

How is the voice input activated and sent to Claude?

Accepted Answer

Voice input is activated using a push-to-talk (PTT) mechanism: press and hold the Right Ctrl key to speak, then release it to transcribe your speech. The resulting text is then automatically injected into your Claude Code session via Tmux.

Question 4

Is Speech-to-Text Linux compatible with other operating systems?

Accepted Answer

No, this tool is exclusively designed for Linux environments. Its functionality relies on direct access to Linux-specific `/dev` devices for keyboard monitoring and audio recording, making it incompatible with other OS platforms.

Question 5

What is Speech-to-Text Linux?

Accepted Answer

Speech-to-Text Linux is a local server that provides push-to-talk voice input functionality for Claude Code on Linux. It transcribes your speech and injects it directly into Claude's conversation stream via Tmux.

Speech-to-Text Linux

关于

主要功能

使用案例