{"id":"W4287591426","doi":"10.48550/arxiv.2011.11588","title":"The Zero Resource Speech Benchmark 2021: Metrics and baselines for unsupervised spoken language modeling","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":42,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canada Research Chairs; University of Toronto","funders":"","keywords":"Computer science; Natural language processing; Benchmark (surveying); Artificial intelligence; Cluster analysis; Spoken language; Language model; Syntax; Speech recognition; Concatenation (mathematics); Pipeline (software); Representation (politics)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005978468,0.0002879392,0.000318881,0.0002403098,0.0003692877,0.0003253608,0.001462918,0.0002152579,0.00002353225],"category_scores_gemma":[0.0004686124,0.0002614928,0.0002305853,0.0007310829,0.00006630334,0.0001823837,0.001291271,0.0003724154,0.00002213099],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006995367,"about_ca_system_score_gemma":0.000110446,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009480162,"about_ca_topic_score_gemma":0.00003199394,"domain_scores_codex":[0.9981572,0.0001622026,0.0002340295,0.0009829529,0.0001356849,0.0003279306],"domain_scores_gemma":[0.9977451,0.0008792916,0.0001441559,0.0008319682,0.0001926276,0.0002068618],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007952626,0.0004164546,0.001614422,0.001028606,0.001337812,0.002101603,0.004542322,0.1129119,0.0007475071,0.1672063,0.01752481,0.689773],"study_design_scores_gemma":[0.0003649668,0.00002641148,0.00001730381,0.00004785386,0.00007974017,0.000004315165,0.0004438222,0.9768644,0.000318306,0.01812886,0.003379793,0.0003242391],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08610871,0.0004624507,0.9093757,0.001060162,0.0002754461,0.0005086386,0.00005930089,0.0001585101,0.001991075],"genre_scores_gemma":[0.965342,0.0009279195,0.03188508,0.0004355251,0.0002089899,0.000004097492,0.000047132,0.00003094707,0.00111829],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8792333,"threshold_uncertainty_score":0.9999837,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0971620928814873,"score_gpt":0.2049767873225302,"score_spread":0.1078146944410429,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}