THIRI × STABLE AUDIO

Your model renders the sound.
THIRI supplies the harmony rail.

Text prompts steer genre, mood, and tempo. Nothing in the prompt layer steers harmony — no key, no chords, no changes. Below is that gap, measured against Stable Audio 3, and what we found trying to close it through init_audio. Press play.

THE DEMO

Same model. Same prompt. Same seed.

Three generations from Stable Audio 3. The text prompt and seed never change — the only variable is a THIRI-computed harmonic skeleton conditioning the render. Switch sources mid-playback and hear the difference at the same bar.

MODEL stable-audio-3 · small-music PROMPT “Smoky late-night jazz trio: warm electric piano chords, upright bass, light brushed drums, 96 BPM” SEED 4242 FORM 12-bar blues in F · 30s
0:00 0:30

Clip B lands on the chart — including the G♭dim7 chromatic passing chord THIRI's reharmonization engine inserted at bar 8 (it shares three of four tones with F7; voice-leading score 0.86).

Clip C moves the harmony without touching the prompt — bars 9–12 swap to the tritone-substitution branch (C7 → G♭7). The vibe stays; the changes move.

WHAT THIS SHOWS — AND WHAT IT DOESN'T. The gap is real and measured: the text-only control grades 2 of 12 bars against the chart, which is chance. Prompting does not hold harmony. But B and C landing on their charts is not evidence the model read one. A pre-registered attribution sweep found accuracy tracks how much of the THIRI skeleton survives diffusion (r = +0.98); at init_noise 0.9, where the skeleton is gone, accuracy falls back to 2 of 12. Both skeletons also grade 12 of 12 alone — so leakage alone reaches the observed ceiling. Audio conditioning is the wrong wire for symbolic control. Read the full negative result →

THE MOMENT

You made generation fast and clean. Musicians will ask for one more thing.

Stable Audio 3 renders in seconds, on licensed data, with editing pros can trust — the hard infrastructure problems, already solved. The next question arrives with every working musician who picks it up: “can it follow my changes?” That's the question this page answers.

PROMPTS

Prompts own the vibe

Genre, mood, instrumentation, tempo — the prompt layer is genuinely good at feel. A chart is a different kind of instruction, and that's the piece we bring: prompts keep the vibe, THIRI holds the changes.

THE PROS

You put musicians at the center

The major-label alliances put professional creators on the roadmap — and pros arrive with changes in hand. THIRI is how the model follows them, demonstrated live above.

THE DAW

You're headed where MIDI lives

The pro surface taking shape around Stable Audio speaks the DAW's language — chord tracks, MIDI, sessions. THIRI is that vocabulary as a service, already running as an API.

You solved speed. You solved trust. Control is the race that's left — and harmonic control is the hardest half of it.

Where THIRI fits
THE ENGINE

Deterministic harmony, served as infrastructure

THIRI is a chord-intelligence engine: analysis, voicings, voice-leading, and reharmonization computed from pitch-class-set theory — not sampled from a language model. It answers the same way every time, in milliseconds, from a production API that agents and plugins can call directly.

Deterministic
Computed, not guessed — C7sus4 keeps its suspension, every time
Live API
Production endpoints for analysis, voicing, reharmonization, and band conduct
Agent-native
Hosted MCP server — Claude, Cursor, or any agent can drive the engine
Skeleton out
Bar-aligned harmonic targets — as symbols or as audio — for conditioning and for training
FOR STABILITY

Four places the rail bolts on

01 · PRO TOOLS

A chord track for the plugin

Any musician-facing Stable Audio surface needs a chord lane the moment a player touches it. THIRI is that lane as a service — suggestion, voicing, reharmonization — without building a theory engine in-house.

02 · OPEN WEIGHTS

A symbolic control branch for SA3

Our own attribution test says the honest version out loud: harmony passed as audio can't be separated from the audio it rode in on. The fix isn't a better skeleton — it's a pathway that accepts chord symbols directly. THIRI supplies the labeled harmonic ground truth to train exactly that.

03 · EDITING

Chord-aware inpainting

Inpainting rebuilds the sound; THIRI context tells each regenerated bar what it must resolve to — so every edit stays harmonically true to the surrounding track.

04 · AGENTS

The theory layer for agent workflows

Agents composing with Stable Audio need harmony they can trust. THIRI's MCP server plans the arc; the model renders it.

We're building this at MUTEK

Our team is registered for the Music Technology Hackathon Montreal, August 22–23 — building on Stable Audio 3. If you're at Stability and want the harmony rail under your model, talk to us before the demo does.

dennison@bluesprincemedia.com · engine live at chords.thiri.ai · MCP at mcp.thiri.ai