Schema | local-whisper-mcp

local-whisper-mcp

Describes the environment variables required to run the server.

Name	Required	Description	Default
`LOCAL_WHISPER_MODEL`	No	Any model from whisper.cpp's model list, by name (base) or file name (ggml-base.bin).	large-v3-turbo-q8_0
`LOCAL_WHISPER_FFMPEG`	No	Full path of the ffmpeg executable. Without it, the server looks on the PATH and in the folders of winget, Scoop, Chocolatey, Homebrew, and MacPorts.	Found automatically
`LOCAL_WHISPER_RUNTIME`	No	auto, cpu, or vulkan. A runtime that can't load falls back to the CPU.	auto
`LOCAL_WHISPER_THREADS`	No	CPU threads for transcription.	Up to 8
`LOCAL_WHISPER_USE_GPU`	No	false runs only on the CPU.	true
`LOCAL_WHISPER_LANGUAGE`	No	A language code such as en or es, or auto to detect it.	auto
`LOCAL_WHISPER_INBOX_DIR`	No	Folder for audio that clients copy to the computer. The server deletes each file there after transcribing it.	~/.local-whisper-mcp/inbox
`LOCAL_WHISPER_MODELS_DIR`	No	Folder for the downloaded models.	~/.local-whisper-mcp/models
`LOCAL_WHISPER_IDLE_MINUTES`	No	Minutes without use before the model leaves memory. 0 frees it after every transcription.	5
`LOCAL_WHISPER_ALLOWED_ROOTS`	No	Folders the server may read audio from, separated by ; on Windows and : elsewhere. The --allowed-roots DIR... option replaces it.	Your user folder
`LOCAL_WHISPER_AUTO_DOWNLOAD`	No	false stops the server from downloading missing models.	true

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Server capabilities have not been inspected yet.

Functions exposed to the LLM to take actions

Name	Description
No tools

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

This server cannot be deployed