{"id":"W4412195314","doi":"10.3390/s25144288","title":"Phoneme-Aware Hierarchical Augmentation and Semantic-Aware SpecAugment for Low-Resource Cantonese Speech Recognition","year":2025,"lang":"en","type":"article","venue":"Sensors","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Science Foundation of Shandong Province; National Natural Science Foundation of China","keywords":"Computer science; Speech recognition; Pronunciation; Word error rate; Context (archaeology); Masking (illustration); Artificial intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.0003634873,0.0008725738,0.0004527734,0.0002435896,0.0002697807,0.0003822946,0.0007378985,0.0003620887,0.001638322],"category_scores_gemma":[0.000936108,0.0002170364,0.0004368263,0.0001962998,0.0004293967,0.0005819338,0.0009673855,0.0008105267,0.0008598714],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002873895,"about_ca_system_score_gemma":0.0007274654,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005847094,"about_ca_topic_score_gemma":0.01234216,"domain_scores_codex":[0.9997284,0.00006871804,0.000009458193,0.00008030904,0.00007006136,0.0000431037],"domain_scores_gemma":[0.999739,0.0000963489,0.00002066528,0.00006845818,0.00005512359,0.00002038414],"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":[0.000568027,0.0001467782,0.001644029,0.0001074896,0.00006292038,0.0002191375,0.0002537531,0.2581974,0.185465,0.00493532,0.004764569,0.5436355],"study_design_scores_gemma":[0.000009465203,0.0001105384,0.0006230129,0.000006975118,0.0000159001,0.00006095564,0.00003245521,0.9614908,0.03331332,0.001851924,0.002467005,0.00001783571],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1002186,0.0005314074,0.8894449,0.0002144724,0.0001266335,0.00005152583,0.0002619762,0.005671429,0.003479099],"genre_scores_gemma":[0.828234,0.0002066413,0.164277,0.0001949797,0.00005892266,0.0001075708,0.0009586762,0.00031418,0.005647996],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005847094,"threshold_uncertainty_score":0.01162612,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01694022235221891,"score_gpt":0.2642564729911856,"score_spread":0.2473162506389667,"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."}}