seed: random-baseline submission (srr_complement, DeepVoice-calibrated)

submissions/random-baseline.yaml

@@ -0,0 +1,40 @@
+schema_version: 5
+system:
+  name: random-baseline
+  slug: random-baseline
+  description: Reference random baseline. Returns N(0,1) for every utterance (seed=0).
+    On a spoof-only set, 1-SRR ~ 50% at the DeepVoice-calibrated threshold by construction.
+  code: https://github.com/SpeechAntiSpoofingBenchmarks/speech-spoof-bench
+  checkpoint: https://huggingface.co/SpeechAntiSpoofingBenchmarks/random-baseline-asas
+  params_millions: 1
+  paper:
+    arxiv_id: '1911.01601'
+    url: https://arxiv.org/abs/1911.01601
+    bibtex: '@article{wang2020asvspoof, title={ASVspoof 2019}, year={2020}}'
+dataset:
+  id: lab260/LRLspoof
+  revision: 5caf11349cf96aa198527cfb040cdba585cdd49e
+  split: test
+scores:
+  srr_complement: 50.01744023366079
+  n_trials: 1304455
+  n_skipped: 0
+calibration:
+  source_dataset: SpeechAntiSpoofingBenchmarks/DeepVoice
+  threshold: 0.0001841719745222704
+artifact:
+  scores_url: https://huggingface.co/SpeechAntiSpoofingBenchmarks/random-baseline-asas/resolve/7c7291672910a9ee755fca6e60e2f1c8a6da7d8a/.eval_results/lab260/LRLspoof/scores.txt
+  scores_sha256: d7a542360024d37ae0ca80cdaca6d1b31f2c9c09f01bad6421235ccc515bcbcd
+  bench_version: speech-spoof-bench==0.4.1
+reproduction:
+  reproduced_by: SpeechAntiSpoofingBenchmarks
+  reproduced_at: '2026-06-21'
+  reproduced_bench_version: speech-spoof-bench==0.4.1
+  match: scoring
+submitter:
+  hf_username: SpeechAntiSpoofingBenchmarks
+  contact: k.n.borodin@mtuci.ru
+submitted_at: '2026-06-21'
+notes: Seeded random baseline for LRLspoof (spoof-only, srr_complement). t* transferred
+  from the random-baseline DeepVoice EER operating point; reproduced via reproduce
+  --scoring (sha matched, value matched).