Datasets:
YAML Metadata Error:Invalid content in eval.yaml.
Check out the documentation for more information.
Show details
✖ Invalid option: expected one of "prompt_template"|"system_message"|"user_message"|"chain_of_thought"|"use_tools"|"generate"|"self_critique"|"multiple_choice"
→ at tasks[0].solvers[0].name
✖ Invalid option: expected one of "includes"|"match"|"pattern"|"answer"|"exact"|"f1"|"model_graded_qa"|"model_graded_fact"|"choice"
→ at tasks[0].scorers[0].name
| name: ADD22_eval_31 | |
| description: > | |
| ADD 2022 Track 3 (Audio Fake Game / FG), Round 1 evaluation partition | |
| (arXiv 2202.08433). Binary anti-spoofing: 20,776 bonafide (genuine) vs 92,085 | |
| spoof (fake), 112,861 utterances total. EER computed over all utterances. | |
| LABELS ONLY: the CC BY-NC-ND 4.0 audio is not redistributed here; only | |
| data/labels.parquet (utterance_id -> label, 0=bonafide / 1=spoof) is shipped, | |
| which is all the arena needs to reproduce a submission's scores. utterance_id | |
| is the audio filename stem, e.g. "ADD_E3_00000000". | |
| evaluation_framework: inspect-ai | |
| tasks: | |
| - id: antispoofing_eval | |
| config: default | |
| split: test | |
| field_spec: | |
| input: audio | |
| target: label | |
| solvers: | |
| - name: speech_spoof_bench_solver | |
| scorers: | |
| - name: speech_spoof_scorer | |
| metrics: | |
| - eer_percent | |