Frequently asked questions

MusiMap is a music intelligence API. We help you analyze, understand, search and recommend audio at scale, from a handful of tracks to multi-million catalogues. The Developer Portal is where you manage your credentials, explore the APIs and monitor your usage.

All core analyses — audio tagging, mood, genre, instrumentation, energy, embeddings, similarity, listener profiling — are powered by our own proprietary deep-learning models, trained in-house on a large, musicologically validated dataset. We do not route audio to third-party foundation models.

Not for audio analysis. Large language models may be used for purely textual helpers such as natural-language search, summarisation or translation. These features are clearly labelled in the documentation and can be disabled on enterprise plans if needed.

Common use cases include music recommendation and similarity search, mood- or genre-based playlisting, sync licensing tools, ad-tech music matching, catalogue enrichment, listener insights and dashboards, and fingerprinting / deduplication workflows.

We accept the most common lossy and lossless formats: MP3, WAV, FLAC, AAC, OGG, M4A, AIFF. If you work with a format not listed here, contact us — we routinely add support on request.

Yes. Heavily compressed files (below ~128 kbps), very short clips (under ~30 seconds), or audio with severe distortion will produce less reliable results. For best accuracy, send full-length tracks at 128 kbps or higher, or lossless formats whenever possible.

Default limits are suitable for typical commercial tracks (up to several hundred megabytes per file). Higher limits are available on enterprise plans. For very large archives, use the batch ingestion endpoint documented in the Web API reference.

Yes. Enterprise customers can grant MusiMap a read-only IAM assume-role on their own Amazon S3 bucket. We pull each file on demand, analyse it and return the result — without ever copying the content into MusiMap’s infrastructure. This is the recommended pattern when you have strict data residency or governance requirements.

No. Audio you submit through the API is used solely to produce the result you requested. It is never fed into our training pipelines, unless you explicitly opt in through a separate written agreement.

Every organisation is provisioned with a dedicated Amazon S3 bucket and its own IAM scope. Content and metadata from one organisation are never shared with, nor accessible to, any other customer. Cross-tenant access is blocked at the bucket policy and IAM level, not only at the application layer.

Primary hosting is on Amazon Web Services, in European Union regions. Additional regions are available on enterprise plans for customers with specific data-residency requirements.

Thirty (30) days after processing by default. You can delete any file at any time through the API. Enterprise plans support custom retention windows, including a zero-retention mode where no audio is ever stored on MusiMap’s side.

Yes. All traffic between your systems and MusiMap uses TLS 1.2 or higher. Data at rest is encrypted with AES-256 managed through AWS KMS. Sensitive fields (API secrets, OAuth tokens) are hashed or encrypted independently of the bucket-level encryption.

Yes. See our Privacy Policy for full details, including legal bases, your rights, data transfers and subprocessors.

MusiMap uses OAuth 2.0. Interactive applications use the Authorization Code flow with refresh tokens; server-to-server integrations use the Client Credentials flow. You register an OAuth client in the Developer Portal, receive a client ID and secret, and obtain access tokens to call the API.

At launch: Python and JavaScript / TypeScript. Java (Android and JVM backends) and Swift (iOS / macOS) will follow shortly after. You can always call the REST API directly from any language.

Yes. Rate limits and usage quotas are published per plan on the Pricing page and enforced with standard HTTP 429 responses. Enterprise plans can negotiate higher sustained and burst limits.

Yes. A separate sandbox base URL lets you develop and test without consuming production quota. Sandbox credentials are available from the Developer Portal once your account is created.

Yes. A free tier with fixed monthly quotas lets you prototype and evaluate the API without a credit card. Paid plans are introduced when you move to production usage. See the Pricing page for exact limits and prices.

Community support and comprehensive documentation are available to all customers. Paid plans include e-mail support with defined response times; enterprise plans add a dedicated Slack channel, solution engineering and a named account manager.

Enterprise plans include a contractual Service Level Agreement covering availability, response times and service credits. The real-time status of the platform is always public on the Status page.