FluentPlay
Six-axis PAD signal profile, across four illustrative practice phases.
3D visualization · illustrative data
PAD · 64/016,001
Isolate stage
Shape · full trajectory
Six axes, four phases, one shape over time.
Each polygon is a complete PAD profile — six axes radiate from a central spine, with vertex distance equal to score × rmax. Stacked across four illustrative phases, the shape changes as the signal profile evolves.
Illustrative data · no clinical claims
Render mode
Options
Axis scores (0 – 1)
All scores illustrative
Drag to rotate · Scroll to zoom
01 · Axes

Six axes, each surfacing a different dimension of pre-articulatory stability.

PAD v1 · Six-dimensional
01 / 06
Acoustic Pressure

Acoustic strain captured across a session. Lower acoustic pressure correlates with more stable motor timing in the pre-articulatory window.

02 / 06
Vocal Steadiness

Prosodic and acoustic steadiness through anticipatory load. Surfaces in the signal before articulation onset.

03 / 06
Onset Consistency

Breath support and phonation onset stability. Without consistent onsets, every other axis compresses — the profile sinks toward center.

04 / 06
Speech Rate

Syllables per minute at stable PAD score. Not raw speed — stability-adjusted speed. Fast and unstable rates score lower than slow and stable.

05 / 06
Session Consistency

Variance across sessions — are profiles returning similar shapes over time? High variance scores low; locked-in practice scores high.

06 / 06
Phoneme Range

Coverage across the phoneme inventory. A speaker can be stable on /m/ and unstable on /p/ — this axis makes that pattern visible.

02 · What PAD measures

Timing instability in the pre-articulatory window, not at articulation.

Most speech tools measure at or after articulation — the audible disfluency events. PAD measures the pre-articulatory window — the motor timing instability in the milliseconds before speech onset. That earlier signal is what makes the six-axis profile possible.

The framework is not a single score. It is a multi-dimensional surface: six axes derived from acoustic, prosodic, and timing features, scored per session, stacked across the session arc. A profile is a shape. A session arc is a shape changing over time.

Clinicians, researchers, and digital therapeutic partners can use the PAD framework as an independent instrument — for signal measurement, for validation of other tools, or as the scoring layer inside their own platform.

PAD score definition
D = P × A × (1 − G · λ)
PAD score = Prediction × Acoustic × (1 − Gate · λ). Full scoring framework detailed in the provisional filing.
Want to see PAD against your own data, or inside your own platform?
Request a meeting → See the audio pipeline → See the platform →
The full picture

The first platform built around the pre-articulatory window.

The audio pipeline captures pre-articulatory timing instability in real time. PAD scores it per syllable. Every game in the FluentPlay library ships with both layers hardwired in. The pipeline and the scoring framework are architecturally independent — licensable separately or as an integrated stack. Patent pending under U.S. Provisional 64/016,001.

How PAD Works
End-to-end platform walkthrough. Start here.
Open walkthrough →
PAD Explainer
Scoring framework deep dive — Prediction, Acoustic, Gate, λ.
Open explainer →
Audio Pipeline
Interactive playback of the analysis engine. Watch features stream from mic to PAD score.
Open pipeline →
EEG Station
Pre-SMA ready-state research rig. fNIRS & EEG alignment.
Open station →
Request a meeting

Tell us what you're working on.

Whether you're evaluating PAD for a clinical platform, a research protocol, or a product integration — describe your use case and we'll schedule a founder call. No obligation, no slide deck.