{"id":"W4414462895","doi":"10.1109/jiot.2025.3613687","title":"Leveraging Cross-Attention Transformer and Multifeature Fusion for Cross-Linguistic Speech Emotion Recognition","year":2025,"lang":"en","type":"article","venue":"IEEE Internet of Things Journal","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Mel-frequency cepstrum; Transformer; Feature extraction; Emotion recognition; Speech processing; Training set; Cepstrum; Acoustic model; Speech corpus","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":[0.0007308869,0.0001400755,0.0001807327,0.0002416522,0.0002303899,0.0008376573,0.0003412653,0.0001115585,0.000008925759],"category_scores_gemma":[0.0002283345,0.0001264343,0.0001294041,0.0001527991,0.00006078759,0.00109524,0.00003188987,0.0003397398,0.000002128096],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007607717,"about_ca_system_score_gemma":0.00005016881,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000188722,"about_ca_topic_score_gemma":0.000001470062,"domain_scores_codex":[0.9988189,0.00002764258,0.0004347966,0.0002676788,0.0002217672,0.0002292428],"domain_scores_gemma":[0.998944,0.0000759722,0.0002874381,0.000102596,0.0005270807,0.0000628811],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001878068,0.00009254162,0.006658489,0.0004196146,0.00007354015,0.00002131181,0.004811251,0.00004520026,0.1914012,0.00007153084,0.0004640532,0.7957535],"study_design_scores_gemma":[0.00251394,0.0002087926,0.01051336,0.00209026,0.00005200849,0.0005845631,0.00008501662,0.03974034,0.9115484,0.03172302,0.0006413661,0.0002989091],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4965997,0.0001277293,0.5014551,0.000215779,0.001240858,0.00007914416,0.000001016427,0.00002611178,0.0002546552],"genre_scores_gemma":[0.9247126,0.00003370968,0.07406623,0.0002957855,0.0001769049,0.000002734466,0.000003361638,0.000008337594,0.0007003288],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7954546,"threshold_uncertainty_score":0.8077548,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02069736530307698,"score_gpt":0.3086427649824087,"score_spread":0.2879453996793317,"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."}}