Temper graph radiation slope

docs/acoustic-graph-flow-audit.md

@@ -135,7 +135,10 @@ The current graph now names each radiation terminal's boundary flow and applies
 a bounded local admittance term derived from that terminal area. This reduces
 the worst over-radiation exposed by the side-branch fix, but it also makes the
 remaining failure clearer: speech-band articulation is still weak, especially
-for utterances. The graph is less loud and still not speaking.
+for utterances. The radiation blend was softened away from a mostly
+differentiated/high-passed flow because the graph was over-bright relative to
+its weak vowel body. That helps some closure/thrombosis evidence but does not
+fix the missing tract resonance.
 
 Required cut: keep moving radiation toward local port admittance and aperture
 physics, but the next audible articulation work probably belongs in pressure

src/AquaSynth.Core/FaustExport.cs

@@ -802,7 +802,7 @@ public static class FaustEmitter
                 var flowName = $"{name}_graph_radiation_flow_{safeTerminal}";
                 source.AppendLine($"    {admittance} = sqrt(clip01(({terminalArea}) / max(0.000001, ({terminalArea}) + 1.0)));");
                 source.AppendLine($"    {flowName} = {flow};");
-                radiated.Add($"(({flowName} * 0.15 + ({flowName} : {filter}) * 0.85) * {opening} * {loss} * {admittance})");
+                radiated.Add($"(({flowName} * 0.30 + ({flowName} : {filter}) * 0.70) * {opening} * {loss} * {admittance})");
             }
         }
 
@@ -873,6 +873,9 @@ public static class FaustEmitter
         var releasePressure = localPressure is null
             ? "1.0"
             : $"(0.20 + 2.80 * clip01((abs({localPressure}) * ({closure})) : + ~ *(0.992)))";
+        var turbulence = $"(no.noise : fi.highpass(2, 900.0 + 2600.0 * clip01(1.0 - {opening})))";
+        var apertureNoiseGate = $"min(clip01((0.8 - {opening}) / 0.8), clip01(8.0 * (0.7 - {opening})) * clip01(30.0 * ({opening} - 0.3)))";
+        var releaseBurstGate = $"clip01({opening} / 0.8) * {release}";
         return port.Kind switch
         {
             AcousticSourceKind.Glottal =>
@@ -882,7 +885,7 @@ public static class FaustEmitter
             AcousticSourceKind.Reed =>
                 $"(ma.tanh(sin(2.0 * ma.PI * {phase}) * (1.0 + {pressure} * 8.0)) * {opening} + no.noise * {noise} * 0.2) * {balance}",
             AcousticSourceKind.TurbulenceJet =>
-                $"((no.noise : fi.highpass(2, 900.0 + 2600.0 * clip01(1.0 - {opening}))) * {noise} * {pressure} * min(clip01((0.8 - {opening}) / 0.8), clip01(8.0 * (0.7 - {opening})) * clip01(30.0 * ({opening} - 0.3))) * 0.62 + {transient} * {pressure} * {release} * {releasePressure} * 0.58 * (0.8 + 0.8 * clip01({opening} / 0.8))) * {balance}",
+                $"(({turbulence}) * {noise} * {pressure} * ({apertureNoiseGate}) * 0.62 + ({turbulence}) * {noise} * {transient} * ({releaseBurstGate}) * 1.2 + {transient} * {pressure} * {release} * {releasePressure} * 0.58 * (0.8 + 0.8 * clip01({opening} / 0.8))) * {balance}",
             AcousticSourceKind.Click =>
                 $"no.noise * {pressure} * max({opening}, {transient}) * exp(0.0 - age * 120.0) * {balance}",
             AcousticSourceKind.Synthetic =>

state/evidence.jsonl

@@ -183,3 +183,4 @@
 {"evidence":"Graph radiation now emits named per-terminal radiation flow/admittance signals and weights boundary flow by bounded local terminal admittance instead of a fixed 4x output scalar plus kind gain. Focused structural and PT parity tests passed. Latest static artifact artifacts/parity/pink-trombone-logmel/20260527T170548495: open/front/nasal/ma/sibilant/closure cosine 0.6675/0.7216/0.7251/0.7098/0.4862/0.6513, rmsRatio 0.2590/0.4287/0.3704/0.4017/0.3957/0.5635. Latest utterance artifact artifacts/parity/pink-trombone-utterance-logmel/20260527T170618466: mama/papa/thrombosis cosine 0.7428/0.7830/0.5442, rmsRatio 2.7955/2.4594/2.7397, articulation 0.1288/0.1226/0.1151.","claim":"Radiation is now less of a global loudness patch and more of a local boundary-flow witness. This reduces over-radiation after the side-port fix, but the speech still fails on articulation, so the next coherent work is passive multi-port scattering plus pressure storage/release rather than more output coloring.","timestamp":"2026-05-27T17:10:00+01:00","topic":"acoustic-graph-radiation-admittance"}
 {"evidence":"Added generated Faust `graph_connection_energy_in_*` and `graph_connection_energy_out_*` probes for the named three-port branch scatter and fenced them with structural tests. Focused structural tests passed.","claim":"Passivity must be inspected where samples are scattered, not inferred from output loudness. The next branch-law pass can now compare local incoming/outgoing area-weighted wave energy before changing more audio behavior.","timestamp":"2026-05-27T17:14:00+01:00","topic":"acoustic-graph-branch-energy-probes"}
 {"evidence":"Moved graph source evaluation into recursive node context by emitting per-node incident pressure and node-local source sums, and moved generated tract injection source banks from graph nodes to cell centers with `1 / sections` index scaling. Focused structural and PT parity tests passed. Latest utterance artifact artifacts/parity/pink-trombone-utterance-logmel/20260527T172949293: mama/papa/thrombosis cosine 0.6654/0.7746/0.5522, rmsRatio 3.0010/2.5712/2.4943, articulation 0.1444/0.1177/0.1193. Latest static artifact artifacts/parity/pink-trombone-logmel/20260527T172843945: closure-release rmsRatio 1.0078, cosine 0.6505.","claim":"Moving turbulence/release sources into graph node context and cell-centered geometry is architecturally correct even though papa articulation still worsens. PT-class source injection is cell-local wave excitation, not a click born on the radiation terminal; the remaining plosive failure points to missing upstream pressure storage rather than source placement.","timestamp":"2026-05-27T17:32:00+01:00","topic":"acoustic-graph-cell-centered-sources"}
+{"evidence":"Added a release-timed burst-noise term to `TurbulenceJet` and softened radiation from an over-bright 15/85 raw/highpassed boundary-flow blend to 30/70 after local admittance weighting. Focused PT parity tests passed. Latest utterance artifact artifacts/parity/pink-trombone-utterance-logmel/20260527T174236049: mama/papa/thrombosis cosine 0.6578/0.7692/0.5585, rmsRatio 3.0006/2.6153/2.3488, articulation 0.1437/0.1173/0.1191. Latest static artifact artifacts/parity/pink-trombone-logmel/20260527T174139661: closure-release rmsRatio 0.9900, speechBandRatio 0.0842.","claim":"Plosive release should include turbulent burst energy, and radiation should not be an almost pure differentiator when tract body is weak. The kept change is modest because stronger source burst did not solve papa; remaining speech failure is tract resonance/scattering and pressure storage, not source loudness.","timestamp":"2026-05-27T17:45:00+01:00","topic":"acoustic-graph-release-burst-radiation-slope"}

state/spine.yaml

@@ -1,6 +1,6 @@
 project: AquaSynth
 spine_version: 1
-updated: 2026-05-27T17:32:00+01:00
+updated: 2026-05-27T17:45:00+01:00
 
 prime_directive: >
   Build coherent synth machinery. References exist to reveal missing structure,
@@ -122,6 +122,7 @@ live_invariants:
   - Pink Trombone log-mel parity now writes graph candidates only. Generated tract graphs mirror live `/pink/...` tract bindings into acoustic source, terminal, radiation, and path-area fields. Graph generation stopped baking first-frame tongue/constriction/lip morphology into the path; `AcousticPath.AreaControl` now turns tongue/constriction/lip controls into live area expressions at graph sample points. Velum owns nasal branch entrance diameter, which lowers through a squared parameter-binding transform into side-port admittance; generated nasal connections keep full connection coupling, and the nasal terminal carries the side-port admittance limit. The first nasal tube sample is no longer allowed to decide the velum opening. Lip opening owns path area and radiation aperture, not terminal area scaling. Graph radiation now uses aperture as a boundary impedance owner: lip/beak opening blends boundary reflection from near-closed termination to declared open-end reflection, names per-port boundary flow (`incoming - outgoing`), and weights output by bounded local terminal admittance rather than a fixed 4x output scalar. Latest static graph evidence after radiation admittance normalization is under `artifacts/parity/pink-trombone-logmel/20260527T170548495`: open/front/nasal/ma/sibilant/closure cosine 0.6675/0.7216/0.7251/0.7098/0.4862/0.6513; latest utterance evidence under `artifacts/parity/pink-trombone-utterance-logmel/20260527T170618466`: mama/papa/thrombosis cosine 0.7428/0.7830/0.5442 with rmsRatio 2.7955/2.4594/2.7397 and articulation 0.1288/0.1226/0.1151. The output is less absurdly loud, but articulation and speech-band motion are still the live failure.
   - `docs/acoustic-graph-flow-audit.md` compares Aqua's graph to Birkholz/VTL, Story/TubeTalker, Smith waveguides, ArtiSynth, and PT/SndKit. The graph records can name the research-backed architecture, but current Faust lowering still mangles multi-port branch sample flow. The kept physical fix moves nasal coupling from global connection damping into transformed side-port admittance and emits a named three-port branch scatter for two same-path ports plus one side branch. Radiation now has named flow/admittance probes and local terminal-admittance weighting, reducing over-radiation without restoring the old damper. The three-port branch law also emits `graph_connection_energy_in_*` and `graph_connection_energy_out_*` probes so passivity can be inspected at the actual scatter site. Do not revert that ownership fix just because the next broken subsystem screams; next cut is passive/admittance-normalized three-port scattering plus graph-native pressure storage/release.
   - Graph source lowering now names per-node incident pressure and node-local source sums inside the recursive graph loop. Generated tract injection source banks now use cell-centered positions and `1 / sections` index scaling, so turbulence/release energy is born inside tube cells rather than directly on graph/radiation nodes. Latest source-position artifact under `artifacts/parity/pink-trombone-utterance-logmel/20260527T172949293`: mama/papa/thrombosis cosine 0.6654/0.7746/0.5522, rmsRatio 3.0010/2.5712/2.4943, articulation 0.1444/0.1177/0.1193. Static closure-release RMS improved to 1.0078 under `artifacts/parity/pink-trombone-logmel/20260527T172843945`, but papa articulation worsened; keep the cell-centered source placement because it is the correct graph-level source geometry, not because metrics liked it.
+  - TurbulenceJet now includes a release-timed burst-noise component in addition to the deterministic pressure pulse, but a stronger trial barely moved papa; the source is not the main remaining articulation bottleneck. Graph radiation now mixes 30% raw boundary flow with 70% high-passed flow after local admittance weighting, down from the over-bright 15/85 blend. Latest radiation/source artifact under `artifacts/parity/pink-trombone-utterance-logmel/20260527T174236049`: mama/papa/thrombosis cosine 0.6578/0.7692/0.5585, rmsRatio 3.0006/2.6153/2.3488, articulation 0.1437/0.1173/0.1191. Static `closure-release` under `artifacts/parity/pink-trombone-logmel/20260527T174139661` has rmsRatio 0.9900 and speechBandRatio 0.0842. This is still not speaking; next pressure is true tract resonance/passive scattering and pressure storage, not more burst gain.
   - PT expressive parity should decompose into reusable Aqua primitives before audio golf: `tract_shape` owns continuous diameter/area geometry over normalized position, physical length (`length_cm`), interpolation/resampling, derived reflection coefficients, and per-cell acoustic delay; `AcousticAreaControl` owns live sampled morphology controls on a path; `WaveClockPolicy` owns the legal delay floor instead of forcing every segment to one sample; generated graph tracts now use `HalfSampleGrid` to match the PT/Kelly-Lochbaum half-sample regime; ordinary same-path graph interiors now own signed two-port area-discontinuity scattering; `glottis` owns excitation controls with mild harmonic waveshaping and a small higher-harmonic brightness term for graph sources; `tract_injection` owns positioned frication/burst controls; `AcousticSourcePort.Transient` owns graph-native release/click energy as a pressure pulse rather than filtered noise; sustained `TurbulenceJet` lowering owns a narrow-open aperture gate so closed constrictions no longer hiss; `AcousticSourcePositionControl` owns differentiable runtime weights over fixed source terminals so moving articulators can move source energy without recompiling the graph; `AcousticRadiationPort` owns aperture-dependent boundary reflection, flow radiation, and current post-flow normalization; `nasal_branch` owns a second diameter/area tube and branch aperture while `AcousticConnection.Coupling` owns structural max admittance; `tract_motion` owns control slew and obstruction threshold; `AcousticPortNetwork` owns graph lowering. `PinkTromboneReferenceRenderer` ports the pinned MIT `chdh/pink-trombone-mod` DSP source (`external/pink-trombone-mod` at 359c2d3b42b10280404c1650dc601902112b4c90), and reference WAVs before `artifacts/parity/pink-trombone-utterances/20260526T211902375` are invalid. The accepted PT utterance curriculum is mama, papa, and thrombosis; lulek was cut because its control sketch taught a bad artifact. Latest graph-only utterance smoke under `artifacts/parity/pink-trombone-utterance-logmel/20260527T150836912`: mama/papa/thrombosis remain `not-accepted-articulation` with cosine 0.8337/0.8893/0.5317 and articulation scores 0.2118/0.1890/0.2594. `AudioComparison.Articulation` reports envelope cosine, active-frame ratio, silence mismatch, envelope/spectral flux ratios, motor-band ratio, speech-band ratio, and an articulation score so log-mel cannot forge a listening receipt. A direct LF coefficient source was tested and cut because it overdrove the graph and broke static open vowel; a broader harmonic-tail source was tested and cut because it worsened voiced-band behavior. A naive midpoint admittance scatter was tested and cut because it collapsed `mama`; the kept area-discontinuity scatter needed the opposite reflection sign for this graph state convention, and re-testing the literal SndKit sign after the clock fix made `thrombosis` loud/unstable. A long noisy release transient was tested and cut because it turned papa into motor-band energy; deterministic pressure release with PT-scale decay is structurally kept but still not sufficient. Next pressure is true graph pressure storage/release at severe closures; output coloring alone is not allowed to impersonate articulation.
   - If dotnet builds hang in this workspace, stop orphan dotnet workers and retry with `--disable-build-servers -p:UseSharedCompilation=false -p:BuildInParallel=false`; plain parallel/shared builds can stall without useful diagnostics.
   - eSpeak NG is an early parity reference for timing, IPA coverage, and log-mel pressure, not an anatomy oracle or final target.