CLI

The AudioLasso CLI is included in the audiolasso npm package.

Install

You can run it without a global install:

npx audiolasso --help

Or install it in a project:

npm install audiolasso

Authenticate

Set an API key:

export AUDIOLASSO_API_KEY="al_..."

Or use browser login:

npx audiolasso login

Separate audio

For a local file:

npx audiolasso separate ./song.wav \
  --prompt "isolate the vocals" \
  --idempotency-key workflow-job-123 \
  --output-dir ./stems

The CLI polls queue status by default while waiting for the result.

For a public URL:

npx audiolasso separate https://example.com/audio.wav \
  --prompt "remove crowd noise" \
  --json

Use --stream when you want live Server-Sent Event status updates instead of polling.

Status and results

Check a job:

npx audiolasso status req_abc123 --logs

Fetch and download a completed result:

npx audiolasso result req_abc123 --download --output-dir ./stems

Cancel queued or running work and release its reserved credits:

npx audiolasso cancel req_abc123 --json

With --json, both successful output and errors are machine-readable JSON for scripts and agents.

MCP

The same package can launch the MCP server:

npx -y audiolasso mcp

Read MCP server for agent setup.