{"id":"W2618015847","doi":"10.21437/interspeech.2017-1178","title":"ASR Error Management for Improving Spoken Language Understanding","year":2017,"lang":"en","type":"preprint","venue":"","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Agence Nationale de la Recherche","keywords":"Computer science; Word error rate; Spoken language; Encoder; Word (group theory); Artificial intelligence; Task (project management); Set (abstract data type); Conditional random field; Natural language processing; Speech recognition; State (computer science); Language model; Machine learning; Algorithm","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":[],"consensus_categories":[],"category_scores_codex":[0.0005915841,0.0002450887,0.000293536,0.0001697485,0.0002445019,0.0009559414,0.001931693,0.0001778166,0.00001077227],"category_scores_gemma":[0.00004829513,0.0002185397,0.0001916012,0.00003961049,0.00002541713,0.0002050139,0.002230007,0.0001730806,0.00004736863],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002407329,"about_ca_system_score_gemma":0.00008549594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004599129,"about_ca_topic_score_gemma":0.0001443108,"domain_scores_codex":[0.9983025,0.00002556071,0.000252359,0.0007392648,0.0002716348,0.0004086978],"domain_scores_gemma":[0.9980361,0.00006539148,0.0002753179,0.001478749,0.00003383569,0.0001106377],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006663767,0.0001190942,0.0003263821,0.00467738,0.0004586155,0.0003336893,0.003898464,0.000171114,0.0003266676,0.8574296,0.01810412,0.1140883],"study_design_scores_gemma":[0.009757099,0.0006057634,0.002613607,0.003412686,0.0005474902,0.00008889035,0.0145559,0.4831472,0.006566626,0.4566914,0.01498145,0.00703187],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0006455356,0.0001995999,0.9199544,0.000379573,0.002993488,0.001102993,0.00001658443,0.0003271217,0.07438068],"genre_scores_gemma":[0.8807134,0.000006803133,0.1087219,0.0001415514,0.0004742132,0.0001895405,0.00003912228,0.00002739923,0.009686042],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8800679,"threshold_uncertainty_score":0.9218165,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07849318842714537,"score_gpt":0.3097299478907526,"score_spread":0.2312367594636072,"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."}}