Using Hosted Models

The models on the Home tab of HARP are hosted by us, as Hugging Face Spaces of the TEAMuP organization, or by Stability AI. They need nothing beyond HARP itself. This page runs one of them, then explains the badges on a model card, API keys, and what to expect from a Space that has been idle.

Separating Stems with Demucs

Demucs separates a mix into drums, bass, vocals, and everything else. It runs on a GPU through ZeroGPU, which Hugging Face lends to each request.

1. Open the model. On the Home tab, search for demucs, or select the Separation category, and click the Demucs card. The first time you open a ZeroGPU model without a Hugging Face key, HARP explains the GPU allowance on which it draws (see API Keys). Click Open to continue.
2. Wait for it to load. The status area shows QUERYING_CONTROLS while HARP reads the interface of the model, then READY. A Space that has been idle takes a minute or two to wake first.
3. Add a song. Drag an audio file (WAV, AIFF, FLAC, OGG, or MP3) onto the Input Audio track, or click its folder icon to choose one.
4. Choose a variant. The Demucs Model dropdown offers htdemucs, which sounds best, mdx_extra, which is faster at a small cost in quality, and mdx_extra_q, a quantized mdx_extra that is the lightest of the three and sounds noticeably worse.
5. Process. Select Process. A typical song takes well under a minute once the Space is awake.
6. Use the stems. The Drums, Bass, Vocals, and Other output tracks fill in as the result arrives. Play them, save them, drag them into your DAW, or drag one into the media clipboard to use as the input of another model, e.g., the vocals into Whisper for a transcript.

Example 2 walks through the code of this model.

Badges

Each model card carries badges, which are worth reading before relying on a model:

• License: the terms of the code and weights of the model, e.g., MIT or CC-BY-NC-4.0. A non-commercial license (NC) means the output cannot be used commercially. Where code and weights are licensed differently, the more restrictive license is shown.
• ZeroGPU: the model runs on a GPU lent to each request, which spends a daily allowance.
• Stability AI: the model runs on the servers of Stability AI, and needs a key from them.
• Availability: Sleeping or Starting for a Space that will take a while to load, and Unavailable, Try again, or Failed to load once a load has failed.

Each model is deployed under its original license, which HARP displays on a badge of its own beside the model, so that the terms are apparent before the output is used. Authors who would rather their model were not featured, or whose license we have stated incorrectly, are asked to open an issue so that we can correct or remove it promptly.

API Keys

Models hosted on Hugging Face run without an account. Those marked ZeroGPU draw on a small daily GPU allowance, which a free Hugging Face token raises. Stability AI models require a Stability AI key, and each request spends credits from that account. Add either under File > Settings > API Keys. HARP sends a key only to the service that issued it, or, for a Hugging Face key, to apps running on your own machine.

The allowance for each kind of account is listed under Plans and Hardware. A Hugging Face key also opens any private Space that its account can see. File > Settings > API Keys links to where each service issues keys, and checks a key when it is added.

Sleeping Spaces and Errors

A Space goes to sleep when nobody has used it for a while, and opening it wakes it. HARP waits while it starts, which takes a minute or two, and the model then loads as usual.

When a request fails, HARP says why in a popup, and offers what helps: adding an API key, opening the page of the model to check its status, opening the log, or reporting the problem through a prefilled GitHub issue. A quota error clears once the allowance recovers, or at once with a key.

Highlights

The Home tab is the full list. A few models to try first:

• Demucs separates a mix into stems.
• Mel-Band RoFormer separates the vocals from the instrumental.
• S-KEY detects the key, as a label over the audio.
• Beat This! tracks beats and downbeats, as labels.
• Transkun transcribes piano audio to MIDI.
• Anticipatory Music Transformer writes an accompaniment to a MIDI file.
• text2midi generates MIDI from a text prompt.
• MusicGen generates music from text, with an optional audio prompt.
• VampNet makes variations on a recording.
• DeepFilterNet denoises speech.
• Whisper transcribes speech to a text file.
• MIDI Synthesizer renders MIDI to audio.

To add a model of your own, see Deploying Models with PyHARP.