{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002764571,0.0001880697,0.0002177063,0.0002565161,0.0002776896,0.0002321157,0.0002208757,0.00009837547,0.00008792536],"category_scores_gemma":[0.000100863,0.0001841718,0.00009139642,0.0003444508,0.00007956765,0.0001729318,0.0000986073,0.0001183412,0.00005652973],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008045878,"about_ca_system_score_gemma":0.00007160169,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005568819,"about_ca_topic_score_gemma":0.0000984666,"domain_scores_codex":[0.9985176,0.0001085262,0.0003000733,0.000516841,0.000249884,0.0003070956],"domain_scores_gemma":[0.9990703,0.0003264622,0.00008182793,0.0002915545,0.0001284427,0.0001014529],"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":[0.00009518095,0.000156193,0.0002011035,0.0001934906,0.0000753391,0.0000413459,0.0006994904,0.00001599718,0.001841927,0.001036246,0.00525999,0.9903837],"study_design_scores_gemma":[0.006608356,0.0003871195,0.006093279,0.001122891,0.0002286021,0.0002123178,0.002738865,0.3643446,0.5396864,0.03575457,0.04109756,0.001725373],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8395795,0.00005901172,0.1458794,0.00718422,0.0007318142,0.001251335,0.00009876091,0.0003842652,0.004831728],"genre_scores_gemma":[0.9092383,0.0001319197,0.07931662,0.003877717,0.0003098316,0.0001928139,0.0002136328,0.0000482604,0.006670928],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9886583,"threshold_uncertainty_score":0.7510308,"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."}}