Audio MCP
An MCP server that provides audio perception capabilities for AI agents, allowing them to measure, analyze, edit, and synthesize sound locally. It supports tasks such as loudness measurement, spectrogram generation, pitch detection, event identification, transcription, and audio comparison without requiring GPU or API keys.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Analyze audio files for loudness, clipping, and noise issues
- Generate spectrograms and waveform visualizations
- Transcribe speech to text using local Whisper
- Edit and synthesize audio through ordered chains
Source and maintenance
- Last repository push
- Latest release
- v0.4.0
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
audio-mcp 0.4.0
{
"version": "0.4.0",
"transport": "stdio",
"identifier": "audio-mcp",
"runtimeHint": "uvx",
"registryType": "pypi",
"environmentVariables": [
{
"name": "AUDIO_MCP_ALLOWED_DIRS",
"description": "Optional sandbox: directories the server may read and write (separated by ';' on Windows, ':' elsewhere). Unset means unrestricted."
},
{
"name": "AUDIO_MCP_MAX_SECONDS",
"description": "Optional: longest section (seconds) loaded into memory per call. Default 1800."
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Maintenance history
Public check and source-refresh events. Editorial and ownership records remain private.
- · repository · passed · metadata changed · updated
- · backfill · recorded · Pre-expansion baseline; unobserved rich metadata remains unknown
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"audio-mcp-mcp": {
"command": "uvx",
"args": [
"audio-mcp==0.4.0"
],
"env": {
"AUDIO_MCP_ALLOWED_DIRS": "<YOUR_AUDIO_MCP_ALLOWED_DIRS>",
"AUDIO_MCP_MAX_SECONDS": "<YOUR_AUDIO_MCP_MAX_SECONDS>"
}
}
}
}Cline
{
"mcpServers": {
"audio-mcp-mcp": {
"command": "uvx",
"args": [
"audio-mcp==0.4.0"
],
"env": {
"AUDIO_MCP_ALLOWED_DIRS": "<YOUR_AUDIO_MCP_ALLOWED_DIRS>",
"AUDIO_MCP_MAX_SECONDS": "<YOUR_AUDIO_MCP_MAX_SECONDS>"
}
}
}
}VS Code
{
"servers": {
"audio-mcp-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"audio-mcp==0.4.0"
],
"env": {
"AUDIO_MCP_ALLOWED_DIRS": "<YOUR_AUDIO_MCP_ALLOWED_DIRS>",
"AUDIO_MCP_MAX_SECONDS": "<YOUR_AUDIO_MCP_MAX_SECONDS>"
}
}
}
}Claude Desktop
{
"mcpServers": {
"audio-mcp-mcp": {
"command": "uvx",
"args": [
"audio-mcp==0.4.0"
],
"env": {
"AUDIO_MCP_ALLOWED_DIRS": "<YOUR_AUDIO_MCP_ALLOWED_DIRS>",
"AUDIO_MCP_MAX_SECONDS": "<YOUR_AUDIO_MCP_MAX_SECONDS>"
}
}
}
}Let your agent find its tools
Find servers like Audio MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.