The home purpose-built for the way you compose.
Cognison is the streaming home for AI-composed music. For artists, that means a venue, a publishing pipeline, and an algorithmic surface that already understands generative output — so the work is tagged, charted, and surfaced as a first-class genre, not as an edge-case footnote.
The default for mainstream DSPs is a single edge-case tag. Cognison treats AI composition as the genre and builds the venue around it.
— 48 dBFS · CREATOR CHAINGeneral streamers tag you. We build the venue around you.
Mainstream DSPs started acknowledging AI-composed music — they did not start ranking it. Cognison starts from the opposite assumption: AI composition is the primary content surface, so every layer of the product is built around it.
AI music = single tag.
Acknowledged as an edge case. Surfaced in one bucket. Discovery still treats the work as an exception to the genre map.
We rank by signal.
Mood/genre ML, model fingerprinting, ranked charts by signal. Provenance as a default, not a footnote.
Tagged from day one.
Upload flow auto-applies toolchain, tempo, mood, and genre tags at ingest — so your work is findable the moment it lands, not weeks after somebody reviews it by hand.
An algorithm that already understands generative output.
Placement driven by the same listen/save signals every DSP reads, just routed through a mood/genre graph that was trained on diffusion and transformer output. The moody ambient piece sits on the moody ambient chart.
Provenance on every release.
Model, prompt chain, and remix lineage are surface-level metadata — not a footnote. Listeners see what made the work without us gating the answer behind a tooltip or a press inquiry.
What you get on signup.
The first screen is a creator upload queue — not a streaming dashboard. Every tool we ship is something an AI musician actually has to use.
A verified artist profile in minutes.
Pick a handle, link your toolchain (Suno, Udio, custom diffusion, in-house), and you have a public creator page with a verified crest on the same day.
Unlimited uploads — stems, drafts, and variants.
Stems, mark-II variants, dub plates. The upload pipeline treats AI output as the primary surface, not a single edge-case tag with a quota.
A built-in analytics dashboard.
Saves, completion rate, listener-fit score, and where you sit on the mood/genre chart — surfaced without a third-party pixel chain and exported as CSV at request.
Public RSS feed for every release.
A real RSS feed on your profile, so listeners can plug you into their reader and follow drops without the platform standing between you and the audience.
No all-in-one mega-stack — just the moves a working AI musician actually has to make on a Tuesday afternoon.
- Genre / mood MLAutomatic, at ingest.
- Model fingerprintingOn every release.
- AI-aware chartsRanked by signal, not by tag.
- Fan followsPer-mood and per-toolchain.
- Provenance metadataModel + prompt chain.
- Creator mapToolchain + mood graph.
Same data as mainstream DSPs. A different model.
We don't penalize AI tracks. We don't artificially bucket them. We read the same listen/save signals every DSP reads — and we route them through a mood/genre graph trained on diffusion and transformer output, so a moody ambient piece sits on the moody ambient chart, not buried under one big “AI Music” bucket.
- SIG.01
Completion rate
How far listeners stay through the track — the same signal every DSP reads.
- SIG.02
Save rate
How often a queue-add is followed by a follow on the artist profile.
- SIG.03
Listener-fit score
How well the mood/genre of this track matches the listener’s recent queue.
- SIG.04
Novelty vs. similarity cluster
How unique the cut is, so we surface a wide variety instead of one mood bucket.
- 01liminal.fm ^ vol.4Udio · custom LoRA · moody · ambient
- 02halflight (demo cut)Suno v4 · prompt chain · downtempo · late-night
- 03tape decay (study)in-house diffusion · lo-fi · dusty groove
We don't bring a moody ambient cut up by accident. We bring it up because the listen pattern — completion, saves, listener-fit — put it there. The chart reflects the audience, not the platform.
Discoverability driven by toolchain + mood overlap.
Not just “people who liked X also liked Y.” A mood/genre map that lets a mood cluster find your new cut before any follower count grows.
AI-music radio surfaces new releases by taste.
Listeners pick the mood, the toolchain, or a starting track — and we surface fresh AI releases that line up with their grid, including cuts from creators with a small-but-tight follow base.
A daily digest of “underdiscovered this week.”
A small subscription surface — opt-in — that pulls the weekly edges of the catalog into one feed. It puts new creators in front of listeners who already care about the genre, not just about following a chart.
Per-follower notifications when you drop.
When you publish, every follower gets a single, opt-in notification — never a feed carpet-bomb. And if a fan follows you inside a mood cluster, they get the mood-cluster drops from other creators in your orbit.
The follower graph is built from toolchain + mood adjacency, not from a single boolean tag. Followers cluster around a mood, not around a platform promise.
What's coming for creators who stake on this.
Subscriptions, membership tiers, and a clean downstream licensing surface — all built so the work can pay for the next cut.
Creator subscriptions.
Tiered drops for serious AI musicians — gated stems, patron-only releases, and process notes for the people who want to see how the prompt chain landed the cut.
Fan memberships.
Listener tiers with early access, lossless playback, and curated crate drops — so the fans that show up first fund the next release instead of waiting for it to trend.
AI licensing for downstream apps.
Opt-in to license your catalog to developers building on top of generative music — a rev-share surface for creators who want their work to live downstream, not just inside the streamer.
Rev-share details are landing before public launch. Joining the waitlist puts you in front of the announcement — and your tools + mood profile are already wired into the surface.
Get a creator profile, a publishing pipeline, and an algorithm that already reads generative output. Cognison starts the moment you upload the first cut.