Music Rights & Licensing

Last updated: 2026-05-25

SoundAssist is a remote collaboration tool for everyone who works with sound; music is the heart of it: artists, producers, engineers (mixing, mastering, recording), labels, A&R, podcasters, sound designers, and the people they share their work with. Uploaded audio is shared with named collaborators or via unguessable share-tokens; it is not a public discovery platform. This distinction matters for what licensing regimes apply.

This page describes what you warrant when you upload content, our position on covers + samples + AI-generated material, and the royalty-collection situation in the relevant jurisdictions.

What you warrant when you upload

By uploading audio to SoundAssist you confirm that, for every track you upload:

You either own the copyright, or have explicit permission from the copyright owner(s) to upload + share the recording.
You have permission to upload any sampled material, interpolations, or other third-party content that the recording contains.
If the recording is a cover version, you have either a mechanical license (e.g. via HFA in the US, BUMA/STEMRA in NL, or the equivalent collecting society) or you are uploading it solely for collaborative work-in-progress and not for public release.
You will not upload content that violates third-party rights including trademark, publicity rights, or confidentiality agreements (NDA leaks, unreleased material from former employers, etc.).

Breach of this warranty entitles SoundAssist to disable access, terminate the account, and pursue damages. Knowingly uploading infringing content is a federal crime under 17 U.S.C. § 506(a) in the US (and equivalent in EU regimes).

Collaboration tool ≠ streaming service

SoundAssist falls under the safe harbor for cloud-storage + collaboration services (US: DMCA § 512(c) + (d); EU: DSA Article 14 hosting safe-harbour). Specifically:

Audio is not indexed on a public discovery surface. Tracks are reachable via share-tokens or by being listed on a user's opt-in public profile.
We do not curate, promote, or monetize specific tracks. Subscription revenue is tier-based (storage, features), not per-stream.
Public profile pages display tracks the user has explicitly marked is_public=true. Everything else is gated.

We are NOT a CSP (Communications Service Provider) for public-performance purposes. If you intend to stream your public-profile tracks to a wide audience, that's your publication decision, and the royalty obligations follow you, not us.

Cover versions + samples

Work-in-progress covers/samples: permitted while the track is private or shared via unguessable token with collaborators. This is "studio work" - the same status as a multitrack session you keep on your hard drive.

Public release of covers: your responsibility. In the US, mechanical compulsory licenses are available via the MLC (musical works) or HFA (private deals). In the Netherlands and EU, BUMA/STEMRA handles mechanicals + performance. Before flipping a cover to is_public=true, obtain the appropriate license.

Public release of sample-based work: sample clearance is required before public release. If you cannot clear a sample, do not make the track public. We may require proof of clearance for tracks that get flagged via DMCA + appear to use copyrighted samples.

AI-generated audio

Generative-AI audio is a rapidly evolving legal area (EU AI Act provisions on AI-generated content begin applying through 2026-2027; US case law is unsettled). Pragmatic position for now:

You may upload AI-generated audio for collaboration + mastering work.
Tracks that contain a substantial AI-generated component and are made public should be labeled as such on the release. We plan to add an "AI-generated" label that uploaders can apply to public tracks.
Using an AI tool that was trained on copyrighted music without permission may itself create rights issues (case law pending). The uploader bears responsibility for understanding the licence of the tool they used.

BUMA/STEMRA + royalty collection (Netherlands)

SoundAssist is registered as a Dutch entity. BUMA/STEMRA (the Dutch performance + mechanical-rights collecting societies) license services that publicly perform copyrighted music.

Our position: as a collaboration tool with private + tokenized sharing as the default, public performance occurs only on opt-in public profile pages and embedded shares. The collected royalty obligation, if any, scales with that public traffic - not with total upload volume.

We will engage BUMA/STEMRA proactively if + when public profile traffic exceeds a threshold that puts us in their tariff structure (current threshold: ~5000 unique listeners/month of public tracks). Until then, individual uploaders who release public tracks are responsible for their own mechanical + neighbouring-rights filings.

Right of publicity + name / likeness

Audio that uses someone's recognisable voice, name, or likeness without permission may violate right-of-publicity laws independently of copyright. Examples: an AI voice clone of a public figure, an uncleared interview clip, a parody that crosses into defamation.

We treat right-of-publicity complaints under the same notice + counter-notice workflow as DMCA (see /legal/dmca), with the additional requirement that the complainant identify the specific likeness/voice claim being made.

When we remove content

See /legal/dmca for the takedown process and /legal/community-guidelines for the enforcement model + appeal process.

Questions

Legal questions (licensing, royalty obligations, B2B contracts): info@soundassist.online.