{"id":"W2902368158","doi":"10.21437/interspeech.2019-2380","title":"Learning Speaker Representations with Mutual Information","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Discriminator; Mutual information; Computer science; Artificial intelligence; Encoder; Speech recognition; Feature learning; Sentence; Joint probability distribution; Feature (linguistics); Pattern recognition (psychology); Natural language processing; Mathematics; Linguistics; Statistics","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0001722311,0.0001443251,0.0001555197,0.0002033591,0.000073399,0.0005358368,0.000485849,0.0001151499,0.0004049513],"category_scores_gemma":[0.00008301789,0.000114561,0.00006677301,0.0001607603,0.00001976679,0.000907658,0.000513955,0.0003872781,0.002531776],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000346113,"about_ca_system_score_gemma":0.0001429834,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006458508,"about_ca_topic_score_gemma":0.00001162273,"domain_scores_codex":[0.9989467,0.00006452786,0.0002297123,0.0002705095,0.0003367728,0.0001517495],"domain_scores_gemma":[0.998924,0.0001053096,0.0001742463,0.0005543573,0.0001840021,0.00005806375],"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":[0.00003533676,0.0001096608,0.008029761,0.0002128273,0.0002620072,0.00002425002,0.008327322,0.02248836,0.0000321357,0.04034531,0.02020441,0.8999286],"study_design_scores_gemma":[0.001161047,0.0002449893,0.0221044,0.0003237899,0.00006390952,0.0001317441,0.002186195,0.8622874,0.003692102,0.004642729,0.1016142,0.001547504],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005273484,0.00000391215,0.7355901,0.0007051911,0.0003132388,0.0002406895,0.000002359822,0.0003415066,0.2575295],"genre_scores_gemma":[0.4995893,0.00004228596,0.4807898,0.001191159,0.0001308131,0.00009314434,0.0001879538,0.000019601,0.01795601],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8983811,"threshold_uncertainty_score":0.9982449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01884537171818267,"score_gpt":0.2462396019807343,"score_spread":0.2273942302625516,"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."}}