voice-scribe

nlink-jp/voice-scribe

★ 0 stars Go AI/LLM Updated today

Local speech-to-text for macOS: whisper.cpp CLI with speaker diarization, and an MCP server for agents that cannot process audio

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "voice-scribe": {
      "command": "go",
      "args": [
        "run",
        "github.com/nlink-jp/voice-scribe@latest"
      ]
    }
  }
}

README Excerpt

Local speech-to-text for macOS: a CLI that transcribes audio with hands the same capability to an agent whose model cannot process audio. No API key. No audio leaves the machine. An agent driven by a model that cannot hear still needs to read recordings. Sending them to a cloud API costs money per minute and puts the audio on someone

Topics

audio-transcriptioncligolanglocal-firstmacosmcpmcp-serverspeaker-diarizationspeech-to-textsrtwebvttwhisperwhisper-cpp

voice-scribe

Quick Install

README Excerpt

Topics

Related AI/LLM Servers