{"id":"W4416581044","doi":"10.1016/j.csl.2025.101907","title":"A robust framework for noisy speech recognition using Frequency-Guided-Swin Transformer","year":2025,"lang":"en","type":"article","venue":"Computer Speech & Language","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Moncton","funders":"","keywords":"Transformer; Robustness (evolution); Convolutional neural network; Word error rate; Pattern recognition (psychology); Deep neural networks; Artificial neural network; Deep learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009314664,0.001055593,0.001195943,0.0008922436,0.0004860144,0.001258796,0.001780788,0.001028363,0.004593009],"category_scores_gemma":[0.001363096,0.0005430847,0.001203798,0.0006182271,0.0006864148,0.00133559,0.001550284,0.001164972,0.003504941],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004959121,"about_ca_system_score_gemma":0.00100824,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003900029,"about_ca_topic_score_gemma":0.00595542,"domain_scores_codex":[0.9992232,0.0001392739,0.00005081493,0.0001686426,0.0003278845,0.00008999404],"domain_scores_gemma":[0.9995754,0.000104061,0.00004021745,0.000102351,0.0001480494,0.00002989831],"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":[0.000702434,0.0001860924,0.0004391551,0.0002168583,0.0001318341,0.000328314,0.000116965,0.1073561,0.2375178,0.04961456,0.004703308,0.5986866],"study_design_scores_gemma":[0.00001462289,0.00008007599,0.0001606889,0.00001133224,0.00003120065,0.0001829266,0.0000207002,0.9416095,0.04529644,0.007413917,0.00515356,0.00002500016],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0008472891,0.00005042122,0.998248,0.0000143675,0.00001430888,0.00001126757,0.00002461824,0.0005518703,0.0002378258],"genre_scores_gemma":[0.1176292,0.0003745622,0.8753018,0.000119764,0.00009529822,0.0001053713,0.0005327177,0.0005972671,0.005244052],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004593009,"threshold_uncertainty_score":0.01536518,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06383290841893033,"score_gpt":0.3086372540781378,"score_spread":0.2448043456592075,"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."}}