io.github.chicogong/ffvoicemedia

ffvoice

Verified · today

Offline speech-to-text & speaker diarization MCP server: transcribe audio on-device, no cloud

Install

claude mcp add ffvoice -- uvx ffvoice

Our take

Local speech-to-text and diarization MCP server built around whisper.cpp, RNNoise, and Python/C++ bindings. The README is unusually explicit that this is an integration layer rather than a new ASR model, which is the right caveat for buyers. First-run model downloads and native wheels/source builds are the operational cost to plan for.

reviewed by hand · 2026-07-06

Something wrong or dead here? Report it

Verification record

last verified: today
github stars: 3
last commit: 2 mo ago
archived: no
license: MIT
in registry since: 2026-05-19

github.com/chicogong/ffvoice-engine