What it does
The server runs Audial's hosted audio tools from any MCP client. Tools cover splitting a track into vocals, drums, bass and other stems, analysing BPM and key, finding song sections, tempo and bars, and adding audio features, tags or timed lyrics to an analysis. Other tools master tracks, build sample packs, convert audio to MIDI, generate music, turn a single isolated sound into an editable Audial Synth or Serum 2 preset, and sing lyrics in a reference voice.
It suits workflows where one track needs several results. An analysis returns an analysis_id that can be passed to audio_features, tag_audio, get_lyrics or generate_samples, so those steps add to it without uploading the file again. The text2vox tool, in per-note mode, fits lyrics that must line up syllable by syllable with a MIDI melody.
Setup requires an Audial account and uv. No API key is needed for interactive use: the first tool call opens a sign-in page in the browser, and the saved sign-in is shared with the audial command line and Python SDK. Claude Desktop users can install a one-click bundle. On servers or in CI, an AUDIAL_API_KEY can be set instead. Tags, lyrics and sample packs wait a few minutes after a new analysis while Audial prepares the stems.
Limitations and requirements
Requires an Audial account and uv. Sign-in must be approved within two minutes, otherwise the tool returns the sign-in link. Servers without an interactive sign-in need an AUDIAL_API_KEY. sound2vital and sound2serum are not for loops or melodies. text2vox needs a single melody line with no chords, and no note may start more than 50 ms before the previous one ends. In per-note mode the token count must equal the note count, the first token must be a syllable, and there is no rest token.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.AudialAI/audial-mcp
Audial audio tools: stems, song analysis, tags, lyrics, mastering, sample packs, MIDI, music, vocals
v0.8.1As shown on registry.modelcontextprotocol.io when we last read it