aria_linear_head
File size: 4,539 Bytes
ab5ae4a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
{
  "model_ver": "v0.2.3.3",
  "refit_level": "L3",
  "g_ver": 3,
  "feature_engine": "native",
  "feature_engine_source": "training/native_features/abi_native.py (numpy/scipy port of abi_praat.py PART 2; no Praat dependency)",
  "label": "cape_v_severity",
  "seg_ver": "stage0-v0.2.1.2",
  "feature_names": [
    "cpps",
    "jitter_local",
    "gne_max",
    "hfno6000",
    "hnrd",
    "h1h2",
    "shimmer_local_db",
    "shimmer_local",
    "psd"
  ],
  "abi_prime_raw": {
    "level": "L3",
    "coefficients": {
      "b0": 12.47,
      "b": [
        -0.5062,
        -0.3403,
        -5.915,
        -1.061,
        0.06068,
        -0.004161,
        4.759,
        -0.1942,
        10.64
      ],
      "lambda": 100.0
    },
    "formula": "b0 + sum_j b_j x_j",
    "abi_orig_coefficients": {
      "intercept": 5.0447730915,
      "scale": 2.9257400394,
      "cpps": -0.172,
      "jitter_local": -0.193,
      "gne_max": -1.283,
      "hfno6000": -0.396,
      "hnrd": 0.01,
      "h1h2": 0.017,
      "shimmer_local_db": 1.473,
      "shimmer_local": -0.088,
      "psd": -68.295
    },
    "clip": [
      0.0,
      10.0
    ]
  },
  "g6_correction": {
    "active": false,
    "weights": [
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0,
      0.0
    ],
    "reason": "G6 finalized under product strict caliber: real_g6_status=FAIL (253 pairs: 209 train / 44 eval); corrector held bypass, identity fallback"
  },
  "g_table": {
    "path": "g_table.json",
    "sha256": "4cc7d521f6989c87b2355f49381081541860c6f60aa9e170f2393d727c8da385"
  },
  "cold_start_thresholds": {
    "T_diag_L0": 3.44,
    "S_diag": 99.87,
    "T_red_cold": 90.0,
    "T_yellow_cold": 75.0,
    "T_yellow_effective_n_lt_5": 75.0,
    "inversion_lock_active": false,
    "reference_share_above_T_diag": 0.0
  },
  "dev_metrics": {
    "target": "PVQD cape_v_severity average / 10 (ABI 0-10 scale)",
    "caliber": "relaxed",
    "holdout_n": 12,
    "holdout_rs": 0.8881118881118882,
    "holdout_mae_abi": 1.254975337236816,
    "holdout_rs_L0": 0.8951048951048951,
    "cv_mean_rs": 0.64148287217087,
    "cv_rs_range": 0.20672451853288648,
    "g3_refit_gate": {
      "rule": "G3 revision R1 (Ryo 2026-10-18): level chosen by CV; hold-out rs >= 0.70; CV rs range <= tau, tau = P95 of the 5-fold range under a rho0 = 0.70 Fisher-z null at the per-fold speaker count",
      "holdout_rs": 0.8881118881118882,
      "holdout_rs_ge_0_70": true,
      "cv_rs_range": 0.20672451853288648,
      "n_per_fold": 42.4,
      "tau": 0.3692486604860881,
      "tau_params": {
        "rho0": 0.7,
        "quantile": 0.95,
        "draws": 1000000,
        "seed": 20261018
      },
      "cv_range_le_tau": true,
      "holdout_vs_L0": {
        "difference": -0.006993006993006867,
        "bootstrap_ci95": [
          -0.2643073593073595,
          0.21900010353574575
        ],
        "bootstrap_valid_draws": 4000,
        "status": "disclosed only; not a gate under ruling (iii)"
      },
      "pass": true
    },
    "g3_refit_gate_original": {
      "rule": "v0.2.1 original: hold-out rs >= 0.70, hold-out rs >= L0 (strict), CV rs range <= 0.10",
      "holdout_rs": 0.8881118881118882,
      "holdout_rs_ge_0_70": true,
      "not_worse_than_L0": false,
      "cv_rs_range": 0.20672451853288648,
      "cv_range_le_0_10": false,
      "pass": false
    }
  },
  "inputs_sha256": {
    "m3_refit": "8bf40e7004188584c845284ca0f72ecb162de63b1131ae230a182505bcba5186",
    "pool_requests": "d2edadb7308c324d60cb9b92d0e1b397ce69994f113a84832b04fda7385e8a20",
    "pool_reports": "a20e9629e0944c7065b4c346d1d08a5bfc98e238c0be16af68a92b18b33ca30a",
    "pool_abi": "6ae7987b79bae861272375d3cd6b782ed232ed87d5d38f4e3e5e0f8942e77b23"
  },
  "release_status": "formal cold-start release; G3 PASS under revision R1; T_diag uncalibrated (pending ruling)",
  "label_change": {
    "from": "cape_v_breathiness (v0.2.0.1)",
    "to": "cape_v_severity",
    "ruling": "Ryo 2026-10-18 item 1"
  },
  "engine_change": {
    "from": "praat-script (v0.2.3.2)",
    "to": "native",
    "basis": "training/native_features/REPORT.md: 9/9 features r >= 0.95 (|r-1| < 1e-12), ABI L0 max |delta| 5.2e-9 over 2,352 endpoints"
  },
  "g3_revision": "R1 (Ryo 2026-10-18): proposal A noise-calibrated tau; ruling (iii) on not-worse-than-L0",
  "t_diag_status": "PLACEHOLDER 3.44 from spec; S_diag and cold red line UNCALIBRATED, pending Ryo ruling",
  "supersedes": "v0.2.3.2 (draft, archived); v0.2.0.1 (archived)"
}