New tagging models and an enterprise-scale REST API provide the structured music understanding modern AI applications depend on.
Berlin, 18 August – Cyanite today announced Tagging 2.0, a major upgrade to its AI music tagging technology. The release introduces new AI tagging models, richer music metadata, and a scalable REST API designed to provide the structured music understanding that modern AI workflows depend on.
Music catalogs are no longer just being searched by people. Increasingly, they’re being explored, organized and acted upon by AI systems. From conversational search and personalized recommendations to catalog automation and intelligent assistants, AI is rapidly changing how music companies interact with their catalogs.
As a result, music metadata is no longer just helping people discover and organize music. It is becoming the foundation that enables AI systems to understand, reason about and work with it.
Music understanding for the AI era
Cyanite sees this shift as a major opportunity for the industry: LLMs are unlocking powerful new ways to work with music, and how well those systems perform depends on the quality of music understanding behind them. Tagging 2.0 strengthens exactly that layer, giving AI systems the structured, reliable metadata that turns music understanding into dependable AI infrastructure.
Built for catalogs of any size, Tagging 2.0 automatically enriches audio with detailed metadata, making catalogs easier to search, organize and work with. It is already being adopted by companies including US music marketplace BeatStars; Dutch distributor The Source; Japanese label Rambling Records; and US sonic branding agency Human Worldwide.
Trusted by companies managing millions of tracks
“Every month, we process hundreds of thousands of new tracks, so both metadata quality and platform reliability are critical. Cyanite’s Tagging 2.0 gives us the music understanding we need to create a better discovery experience for our creators.” – Anirudh Mani, Chief Science Officer at BeatStars
Tagging 2.0 introduces several new music understanding models, including Lyric Analysis, Tempo, Vocal Style, Music For/Use Case and Structural Segmentation. Existing models such as BPM and Instruments have also been significantly improved.
In addition, tagging granularity has increased from 15-second to 10-second segments, generating up to 2,500 data points per minute of music for a more detailed understanding of every track.
Music understanding becomes AI infrastructure
Together, these improvements provide AI systems with more detailed and more reliable understanding about music, enabling everything from conversational search and recommendation to intelligent catalog management.
The release is accompanied by a new REST API built for higher throughput and scalability, allowing customers to process catalogs ranging from thousands to millions of tracks.
“AI is changing how the music industry works with music catalogs. But these new workflows are only as good as the music understanding behind them. With Tagging 2.0, we’re strengthening that understanding layer – the foundation that makes intelligent music workflows possible.” – Markus Schwarzer, CEO & Co-Founder at Cyanite
Beyond metadata enrichment, Tagging 2.0 also enhances Cyanite’s Free Text Search and Similarity Search by adding richer musical context to every search result, making recommendations more transparent while providing the structured understanding AI-powered applications increasingly rely on.
Today, more than 200 music companies – including BeatStars, Epidemic Sound, Warner Chappell, LANDR and APM Music – use Cyanite to power search, discovery and music understanding across their catalogs.
About Cyanite
Cyanite is an AI music understanding platform that helps streaming services, publishers, distributors, and music technology companies enrich, search, and organize their music catalogs. Its APIs analyze audio directly to generate structured music metadata and power AI-driven search, discovery, and recommendation experiences. Cyanite has analyzed more than 55 million songs and is trusted by more than 200 companies worldwide, including Warner Chappell, BeatStars, Epidemic Sound, LANDR, and APM Music.
Media contact
Jakob Höflich
CMO at Cyanite
jakob@cyanite.ai
For interview requests or additional data, please contact: jakob@cyanite.ai