{"id":"W4389524018","doi":"10.18653/v1/2023.emnlp-main.182","title":"Generative Spoken Language Model based on continuous word-sized audio tokens","year":2023,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Grand Équipement National De Calcul Intensif; Agence Nationale de la Recherche; Canadian Institute for Advanced Research","keywords":"Computer science; Natural language processing; Language model; Word (group theory); Speech recognition; Artificial intelligence; Generative grammar; Spoken language; Generative model; Bridging (networking); Linguistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003802278,0.0005830365,0.0005818634,0.0003259158,0.000185866,0.0007526004,0.001245931,0.0007452158,0.005278244],"category_scores_gemma":[0.001604694,0.0003446167,0.0008200614,0.0003275181,0.0005752518,0.0009501888,0.0007012763,0.001001166,0.00238428],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004502168,"about_ca_system_score_gemma":0.0005620878,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002680674,"about_ca_topic_score_gemma":0.003158834,"domain_scores_codex":[0.9996666,0.00009476329,0.00002039369,0.0001127131,0.00007316213,0.00003245348],"domain_scores_gemma":[0.9992969,0.0004421281,0.0000440837,0.00006917003,0.0001173417,0.00003041811],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003669696,0.00008797651,0.0008461648,0.0002700811,0.0001025193,0.0004682935,0.0004387399,0.7924283,0.0420756,0.03703134,0.00393882,0.1219453],"study_design_scores_gemma":[0.00001060831,0.00002968625,0.00007464809,0.000005103505,0.00001143015,0.00004911071,0.00001043606,0.9927198,0.002370115,0.004004211,0.0007061171,0.000008718275],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02471211,0.000323157,0.9687262,0.0002650127,0.0001325924,0.00006775004,0.0004538962,0.002637844,0.00268145],"genre_scores_gemma":[0.8012351,0.0004546027,0.1763713,0.0003256572,0.0001201553,0.0004202882,0.001311113,0.0005105543,0.01925122],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005278244,"threshold_uncertainty_score":0.01765746,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02701431847517769,"score_gpt":0.2647048678193151,"score_spread":0.2376905493441374,"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."}}