How your session is built — from the moment you tell it what you need, to the moment you press play.

Stage 01 — Intake

What you said, read into a profile.

You open the wizard, or you speak into it. You pick a listening context — bedtime, meditation, focused rest, active work, background, care-companion. You pick a length, anywhere from twenty minutes to two hours. You confirm whether headphones are in. You pick one to three support areas from a vocabulary of thirty-seven. Note the wording: “quiet racing thoughts”, not “treat anxiety”. “Dementia comfort”, not “dementia therapy”. And you can describe the moment in your own words.

Underneath, a language model on your own device reads what you wrote into a structured profile — the choices that drive composition. A safety check runs over the same input, looking for signs of crisis; anything flagged routes you to support resources rather than to a session. The choices are also run through a compatibility check, so non-fit pairings (the bare bed and trauma-aware support, for example) are surfaced before the renderer starts.

Stage 02 — Generation

A session written for the next hour you're in.

The composition planner reads your profile and lays out the session — a plan for every minute. If headphones are in, the session opens with a gateway phase. Two slightly different tones, one to each ear, descend over seven to twelve minutes from a beta entry state to the target band: alpha for relaxation, theta for meditation, delta for sleep onset. The binaural component sits just audible beneath an ambient bed. The architecture borrows from the Monroe Institute's Hemi-Sync method as design lineage, not as clinical claim.

The body is recipe-driven sound: warm harmonic drones, slow breath-paced pads, nature textures, spatial movement, low rumble where it helps, sparse motifs where attention needs a hand. The close brings you back to ordinary alertness — or, if it's bedtime, hands you to sleep without a wake-up. Before the file reaches you, every render is checked for stereo separation, spectral richness, transition smoothness, safe peaks, and conservative high-frequency energy across the length of the session.

Stage 03 — Listening

What you take away from the session.

You listen in your own time. The mobile player shows the current phase, takes lock-screen controls, runs in the background, and resumes where you left off. The compressed download is a portability copy with intro and cue burn-ins; clean unwatermarked playback lives in the app under the base subscription.

Stage 04 — Reflection

A journal that sits next to the listening, if you want one.

Listening is the work. A reflection on the listening is a sidecar — never a streak, never a prompt, never a guilt loop. The journal lives next to every recording in your library, in three short modes you can use or ignore.

Pre-session.A sentence or two about what you're bringing in, captured before you press play. The recording shows the linked entry on its detail page, so you can read what you wrote going in.

Post-session.What changed, or didn't, after the audio finished. Same short-form note. The two together build the simplest possible practice loop without asking you to perform reflection.

Standalone.A reflection that isn't tied to a session at all — for the days when you want to write something down and there's no audio in between.

The journal is yours alone. Body text never leaves your account unless you opt into the optional Insights surface later — and even then only structured fields (entry type, stress / energy 1–10 sliders) are aggregated. The body text of your entries is never read by the analytics path.

The shape of a session

Why a session walks an arc, not a flat line.

The body of every Soundscaper session is built as an ordered chain of phases with short bridges between them. Sequencing and pacing carry weight in clinical music therapy — the order of techniques, and the way each is led into and out of, is itself part of the work.

Short sessions collapse to fewer phases — but never to a single block. Long sessions get the full chain. The composition view in the wizard shows which phases your length supports as you choose. The frequency-activity chart, also in the view, shows how the active technique shifts across the body, so you can see what changes when — not just that something is happening.

What you'll actually hear

Five ways the same session can sound.

What sits underneath a session can be packaged five ways. Each delivers the same underlying composition at the same specification — what changes is the surface you hear. You choose. The wizard suggests a default for your support area; you can override it.

How perceptible the technique is

How visible the technique is, dialled by you.

Inside whichever surface you've chosen, the technique can sit in one of three positions. The underlying composition is the same at all three; what changes is whether you can consciously hear it.

1. Carrier-led (default)

   The techniques are at full specification but woven into the surface, so a listener focused on the music or the nature wouldn't notice them as separate elements. The modulation is at a depth that influences the body's response without reading as audible tremolo. The headphone induction is perceptible as effect, not as two separate tones. Transitions cross-fade. The technique works; you get the listening.

2. Technique-perceptible

   The same techniques, a half-step more visible. The headphone induction is faintly audible as two tones before they fuse. Modulation depth lifts a little, so the rhythmic envelope on a music layer is detectable. Transitions between phases run longer and read more audibly. For listeners who want to feel the technique working.

3. Technique-only

   Available only with the bare bed. The technique is the entire listening experience: drone at full level, breath-paced amplitude clearly present, headphone induction audible as a perceived beat, phase transitions read as deliberate moves. Never a default. Always a deliberate choice.

The guardrails

The audio will never break the listening state.

The defining failure mode for long-form wellness audio is the moment that breaks the listening state — a sudden tonal shift, an unintended transient, a key change without a pivot, a high-frequency partial that reads as harsh. One jarring moment undermines every other design decision in the session. So we enforce a set of engineering invariants: minimum cross-fade durations on every layer transition, dynamic-range ceilings at phase boundaries, high-frequency limits for long sessions, smooth tonal continuity that prevents abrupt key changes, transient checks on every render. None of this is visible to you. It's just absent.

The full set of invariants lives in the codebase and the methods document; you don't have to read them. What matters is the experience: nothing in a Soundscaper session should jolt you, surprise you, or break your attention. If something does, it's a bug, not a feature, and we want to know about it.

What runs where

The work happens on your own machine.

The evidence position

Established. Emerging. Speculative.

Every claim on this site is graded into one of three tiers, and the tier is visible wherever the claim appears. We reject solfeggio frequencies, “DNA repair” claims, the Mozart Effect, and the wellness habit of inflating thin findings. We’re open about the design lineage of the headphone induction phase — Monroe Institute, Hemi-Sync, Gateway — without inheriting the consciousness claims.