binwang commited on
Commit
2cf7728
·
verified ·
1 Parent(s): 5f0ceb5

Upload folder using huggingface_hub

Browse files
results_organized/bleu/st.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,37.60224687716629,43.941098854450516,14.40739
4
  hy_whisper_local_cs,1.0869208512565696,0.10573269629215352,0.008950516549431693,22.267131378964944,7.31707791416422,2.8610263518826757
5
  Qwen2-Audio-7B-Instruct,16.325186897428104,25.765420247070075,0.03245972071872916,6.326113431899141,16.466557744958333,0.04425838146050298
6
  whisper_large_v3,1.600581653970121,0.16408986541757878,0.02107778621423822,46.01512198258627,14.673689493155793,2.451098639578599
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,27.620150160643625,35.274306071307024,8.433062902024755,46.80524126004861,15.209998552437538,2.8327095799289337
8
  gemini-1.5-flash,,,,,,
9
  WavLLM_fairseq,13.841886973016162,31.96381187282953,0.0033159224040994286,5.933522277713613,2.368659001743569,0.1695522548322915
 
4
  hy_whisper_local_cs,1.0869208512565696,0.10573269629215352,0.008950516549431693,22.267131378964944,7.31707791416422,2.8610263518826757
5
  Qwen2-Audio-7B-Instruct,16.325186897428104,25.765420247070075,0.03245972071872916,6.326113431899141,16.466557744958333,0.04425838146050298
6
  whisper_large_v3,1.600581653970121,0.16408986541757878,0.02107778621423822,46.01512198258627,14.673689493155793,2.451098639578599
7
+ old_models,,,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,27.620150160643625,35.274306071307024,8.433062902024755,46.80524126004861,15.209998552437538,2.8327095799289337
9
  gemini-1.5-flash,,,,,,
10
  WavLLM_fairseq,13.841886973016162,31.96381187282953,0.0033159224040994286,5.933522277713613,2.368659001743569,0.1695522548322915
results_organized/llama3_70b_judge/accent_recognition.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,47.01682396389003,7.816666666666666,77.833333
4
  hy_whisper_local_cs,,,
5
  Qwen2-Audio-7B-Instruct,29.187525646286417,2.55,0.9666666666666667
6
  whisper_large_v3,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,24.640951990151827,26.016666666666666,7.633333333333334
8
  gemini-1.5-flash,,,
9
  WavLLM_fairseq,39.96717275338531,2.6833333333333336,0.23333333333333336
 
4
  hy_whisper_local_cs,,,
5
  Qwen2-Audio-7B-Instruct,29.187525646286417,2.55,0.9666666666666667
6
  whisper_large_v3,,,
7
+ old_models,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,24.640951990151827,26.016666666666666,7.633333333333334
9
  gemini-1.5-flash,,,
10
  WavLLM_fairseq,39.96717275338531,2.6833333333333336,0.23333333333333336
results_organized/llama3_70b_judge/audio_captioning.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,38.00454545454545,33.97687861271676
4
  hy_whisper_local_cs,,
5
  Qwen2-Audio-7B-Instruct,40.77727272727273,33.78034682080925
6
  whisper_large_v3,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,3.0954545454545457,6.3468208092485545
8
  gemini-1.5-flash,,
9
  WavLLM_fairseq,5.5,6.901734104046243
 
4
  hy_whisper_local_cs,,
5
  Qwen2-Audio-7B-Instruct,40.77727272727273,33.78034682080925
6
  whisper_large_v3,,
7
+ old_models,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,3.0954545454545457,6.3468208092485545
9
  gemini-1.5-flash,,
10
  WavLLM_fairseq,5.5,6.901734104046243
results_organized/llama3_70b_judge/audio_scene_question_answering.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,63.15021876519203,49.77635782747604,46.315789
4
  hy_whisper_local_cs,,,
5
  Qwen2-Audio-7B-Instruct,50.919591292758774,45.75079872204473,44.473684210526315
6
  whisper_large_v3,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,24.647544968400585,18.466453674121407,18.88157894736842
8
  gemini-1.5-flash,,,
9
  WavLLM_fairseq,43.01199466903598,29.840255591054312,26.25
 
4
  hy_whisper_local_cs,,,
5
  Qwen2-Audio-7B-Instruct,50.919591292758774,45.75079872204473,44.473684210526315
6
  whisper_large_v3,,,
7
+ old_models,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,24.647544968400585,18.466453674121407,18.88157894736842
9
  gemini-1.5-flash,,,
10
  WavLLM_fairseq,43.01199466903598,29.840255591054312,26.25
results_organized/llama3_70b_judge/emotion_recognition.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,48.505976095617534,46.206896551724135,36.3601
4
  hy_whisper_local_cs,,,
5
  Qwen2-Audio-7B-Instruct,53.98406374501992,53.9463601532567,41.60919540229885
6
  whisper_large_v3,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,44.322709163346616,56.59003831417625,47.356321839080465
8
  gemini-1.5-flash,,,
9
  WavLLM_fairseq,59.76095617529881,51.072796934865906,41.57088122605364
 
4
  hy_whisper_local_cs,,,
5
  Qwen2-Audio-7B-Instruct,53.98406374501992,53.9463601532567,41.60919540229885
6
  whisper_large_v3,,,
7
+ old_models,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,44.322709163346616,56.59003831417625,47.356321839080465
9
  gemini-1.5-flash,,,
10
  WavLLM_fairseq,59.76095617529881,51.072796934865906,41.57088122605364
results_organized/llama3_70b_judge/gender_recognition.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,99.75379565038982,93.48605577689243,66.133333
4
  hy_whisper_local_cs,,,,
5
  Qwen2-Audio-7B-Instruct,99.1177677472302,92.80876494023903,68.38333333333333,61.56666666666667
6
  whisper_large_v3,,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,34.94050061551087,15.737051792828685,26.35,19.6
8
  gemini-1.5-flash,,,,
9
  WavLLM_fairseq,69.61427985227739,51.932270916334666,49.06666666666666,46.766666666666666
 
4
  hy_whisper_local_cs,,,,
5
  Qwen2-Audio-7B-Instruct,99.1177677472302,92.80876494023903,68.38333333333333,61.56666666666667
6
  whisper_large_v3,,,,
7
+ old_models,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,34.94050061551087,15.737051792828685,26.35,19.6
9
  gemini-1.5-flash,,,,
10
  WavLLM_fairseq,69.61427985227739,51.932270916334666,49.06666666666666,46.766666666666666
results_organized/llama3_70b_judge/music_understanding.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,57.7927548441449
4
  hy_whisper_local_cs,
5
  Qwen2-Audio-7B-Instruct,71.60909856781802
6
  whisper_large_v3,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,51.727042965459134
8
  gemini-1.5-flash,
9
  WavLLM_fairseq,44.3133951137321
 
4
  hy_whisper_local_cs,
5
  Qwen2-Audio-7B-Instruct,71.60909856781802
6
  whisper_large_v3,
7
+ old_models,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,51.727042965459134
9
  gemini-1.5-flash,
10
  WavLLM_fairseq,44.3133951137321
results_organized/llama3_70b_judge/sds_singlish.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,48.4,46.4,57.0,62.599999999999994
4
  hy_whisper_local_cs,,,,
5
  Qwen2-Audio-7B-Instruct,33.8,24.8,40.4,46.2
6
  whisper_large_v3,,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,45.4,44.0,58.0,65.4
8
  gemini-1.5-flash,,,,
9
  WavLLM_fairseq,31.6,31.6,45.199999999999996,49.400000000000006
 
4
  hy_whisper_local_cs,,,,
5
  Qwen2-Audio-7B-Instruct,33.8,24.8,40.4,46.2
6
  whisper_large_v3,,,,
7
+ old_models,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,45.4,44.0,58.0,65.4
9
  gemini-1.5-flash,,,,
10
  WavLLM_fairseq,31.6,31.6,45.199999999999996,49.400000000000006
results_organized/llama3_70b_judge/speech_instruction.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,65.6,74.80000000000001
4
  hy_whisper_local_cs,,
5
  Qwen2-Audio-7B-Instruct,44.800000000000004,52.599999999999994
6
  whisper_large_v3,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,72.2,73.8
8
  gemini-1.5-flash,,
9
  WavLLM_fairseq,19.2,21.6
 
4
  hy_whisper_local_cs,,
5
  Qwen2-Audio-7B-Instruct,44.800000000000004,52.599999999999994
6
  whisper_large_v3,,
7
+ old_models,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,72.2,73.8
9
  gemini-1.5-flash,,
10
  WavLLM_fairseq,19.2,21.6
results_organized/llama3_70b_judge/sqa_english.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,86.76470588235293,59.7093023255814,73.6647355
4
  hy_whisper_local_cs,,,,,
5
  Qwen2-Audio-7B-Instruct,80.04901960784315,58.31395348837209,64.86264249672958,74.7247908410392,66.49242028227914
6
  whisper_large_v3,,,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,88.57843137254902,73.11046511627907,88.61894972902262,91.85380889476001,89.33612127548353
8
  gemini-1.5-flash,,,,89.25583443416997,
9
  WavLLM_fairseq,83.92156862745098,58.54651162790698,77.64903756307233,66.31439894319684,66.5446941975954
 
4
  hy_whisper_local_cs,,,,,
5
  Qwen2-Audio-7B-Instruct,80.04901960784315,58.31395348837209,64.86264249672958,74.7247908410392,66.49242028227914
6
  whisper_large_v3,,,,,
7
+ old_models,,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,88.57843137254902,73.11046511627907,88.61894972902262,91.85380889476001,89.33612127548353
9
  gemini-1.5-flash,,,,89.25583443416997,
10
  WavLLM_fairseq,83.92156862745098,58.54651162790698,77.64903756307233,66.31439894319684,66.5446941975954
results_organized/llama3_70b_judge/sqa_singlish.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,51.4,53.2,64.80000000000001,67.2
4
  hy_whisper_local_cs,,,,
5
  Qwen2-Audio-7B-Instruct,42.0,39.6,51.6,53.6
6
  whisper_large_v3,,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,56.0,66.0,74.0,71.6
8
  gemini-1.5-flash,,,,
9
  WavLLM_fairseq,45.199999999999996,46.6,50.8,62.199999999999996
 
4
  hy_whisper_local_cs,,,,
5
  Qwen2-Audio-7B-Instruct,42.0,39.6,51.6,53.6
6
  whisper_large_v3,,,,
7
+ old_models,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,56.0,66.0,74.0,71.6
9
  gemini-1.5-flash,,,,
10
  WavLLM_fairseq,45.199999999999996,46.6,50.8,62.199999999999996
results_organized/llama3_70b_judge/under_development_llama3_70b_judge.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,64.51231527093596,53.97524071526823,40.970695
4
  hy_whisper_local_cs,60.137931034482754,60.9353507565337,45.78754578754578
5
  Qwen2-Audio-7B-Instruct,60.453201970443345,51.5818431911967,36.97802197802198
6
  whisper_large_v3,,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,70.18719211822659,64.12654745529574,55.01831501831502
8
  gemini-1.5-flash,78.06896551724138,65.9697386519945,49.908424908424905
9
  WavLLM_fairseq,60.70935960591133,55.625859697386524,40.95238095238095
 
4
  hy_whisper_local_cs,60.137931034482754,60.9353507565337,45.78754578754578
5
  Qwen2-Audio-7B-Instruct,60.453201970443345,51.5818431911967,36.97802197802198
6
  whisper_large_v3,,,
7
+ old_models,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,70.18719211822659,64.12654745529574,55.01831501831502
9
  gemini-1.5-flash,78.06896551724138,65.9697386519945,49.908424908424905
10
  WavLLM_fairseq,60.70935960591133,55.625859697386524,40.95238095238095
results_organized/meteor/audio_captioning.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,0.24920047034353812,0.3175511907248581
4
  hy_whisper_local_cs,,
5
  Qwen2-Audio-7B-Instruct,0.19891712076314283,0.21342294856199182
6
  whisper_large_v3,,
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.05796819723943051,0.120421856260385
8
  gemini-1.5-flash,,
9
  WavLLM_fairseq,0.041732965094428545,0.06399522524688675
 
4
  hy_whisper_local_cs,,
5
  Qwen2-Audio-7B-Instruct,0.19891712076314283,0.21342294856199182
6
  whisper_large_v3,,
7
+ old_models,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.05796819723943051,0.120421856260385
9
  gemini-1.5-flash,,
10
  WavLLM_fairseq,0.041732965094428545,0.06399522524688675
results_organized/wer/asr_english.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,0.022918474365262006,0.041576030415949455,0.0
4
  hy_whisper_local_cs,0.02554042328441544,0.053417065466169825,0.1066766923091754,0.1991585778678581,0.0948233719154953,0.10871196540338629,0.1463228189913085,0.0467690997480572,0.05275660343910654
5
  Qwen2-Audio-7B-Instruct,0.035141660693401744,0.060415760304159495,0.11438872500819404,0.2165498391593041,0.11723812890302816,0.18872219319407232,0.23542555661330924,0.06114048472375004,0.08739585179932637
6
  whisper_large_v3,0.01878749009695552,0.03660128246354058,0.10001863741235596,0.14602420615337386,0.09459022434812692,0.11863959266711877,0.15887899737116104,0.037649480146197796,0.03208650948413402
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.032349945297468596,0.05307658841999735,0.10600831614192711,0.20140159998943682,0.09948381629977261,0.11416493424197618,0.1448629161356777,0.04900464852205386,0.04396383619925545
8
  gemini-1.5-flash,,,,,,,,,
9
  WavLLM_fairseq,0.02103218017882069,0.04798834811886432,0.14533325621300636,0.3792176325635977,0.15491778414546403,0.6447482518259942,0.6671766188447099,0.06621482559171073,0.4536784258110264
 
4
  hy_whisper_local_cs,0.02554042328441544,0.053417065466169825,0.1066766923091754,0.1991585778678581,0.0948233719154953,0.10871196540338629,0.1463228189913085,0.0467690997480572,0.05275660343910654
5
  Qwen2-Audio-7B-Instruct,0.035141660693401744,0.060415760304159495,0.11438872500819404,0.2165498391593041,0.11723812890302816,0.18872219319407232,0.23542555661330924,0.06114048472375004,0.08739585179932637
6
  whisper_large_v3,0.01878749009695552,0.03660128246354058,0.10001863741235596,0.14602420615337386,0.09459022434812692,0.11863959266711877,0.15887899737116104,0.037649480146197796,0.03208650948413402
7
+ old_models,,,,,,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.032349945297468596,0.05307658841999735,0.10600831614192711,0.20140159998943682,0.09948381629977261,0.11416493424197618,0.1448629161356777,0.04900464852205386,0.04396383619925545
9
  gemini-1.5-flash,,,,,,,,,
10
  WavLLM_fairseq,0.02103218017882069,0.04798834811886432,0.14533325621300636,0.3792176325635977,0.15491778414546403,0.6447482518259942,0.6671766188447099,0.06621482559171073,0.4536784258110264
results_organized/wer/asr_mandarin.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,0.13165449110094832
4
  hy_whisper_local_cs,0.16361782582011838
5
  Qwen2-Audio-7B-Instruct,0.09260359129694522
6
  whisper_large_v3,0.12359684029221357
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.20886539565639167
8
  gemini-1.5-flash,
9
  WavLLM_fairseq,0.7054601967888183
 
4
  hy_whisper_local_cs,0.16361782582011838
5
  Qwen2-Audio-7B-Instruct,0.09260359129694522
6
  whisper_large_v3,0.12359684029221357
7
+ old_models,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.20886539565639167
9
  gemini-1.5-flash,
10
  WavLLM_fairseq,0.7054601967888183
results_organized/wer/asr_singlish.csv CHANGED
@@ -4,6 +4,7 @@ MERaLiON-AudioLLM-Whisper-SEA-LION,0.042254894789457,0.048088629169710254,0.2919
4
  hy_whisper_local_cs,0.06319947333772219,0.2719340962584206,0.23856138159502538,0.33742408429629445,0.16663991478309087,0.12873269917149824
5
  Qwen2-Audio-7B-Instruct,0.07197717796796138,0.1905689473257041,0.35076166942732234,0.5613424034000176,0.27856006770658537,0.2245352799625317
6
  whisper_large_v3,0.06844171360300393,0.3171008846684522,0.27026366524560785,0.4618189591218298,0.2143555471246589,0.1698509342851144
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.07041669714480775,0.32988393799204613,0.3035544573275043,0.4779640131272869,0.22881615619208825,0.1789273082575623
8
  gemini-1.5-flash,,,,,,
9
  WavLLM_fairseq,0.10077292565771828,0.4463923382842302,0.7540934640345399,1.143645714142011,0.39796588405247263,0.42541061709652933
 
4
  hy_whisper_local_cs,0.06319947333772219,0.2719340962584206,0.23856138159502538,0.33742408429629445,0.16663991478309087,0.12873269917149824
5
  Qwen2-Audio-7B-Instruct,0.07197717796796138,0.1905689473257041,0.35076166942732234,0.5613424034000176,0.27856006770658537,0.2245352799625317
6
  whisper_large_v3,0.06844171360300393,0.3171008846684522,0.27026366524560785,0.4618189591218298,0.2143555471246589,0.1698509342851144
7
+ old_models,,,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.07041669714480775,0.32988393799204613,0.3035544573275043,0.4779640131272869,0.22881615619208825,0.1789273082575623
9
  gemini-1.5-flash,,,,,,
10
  WavLLM_fairseq,0.10077292565771828,0.4463923382842302,0.7540934640345399,1.143645714142011,0.39796588405247263,0.42541061709652933
results_organized/wer/under_development_wer.csv CHANGED
@@ -1,11 +1,12 @@
1
  Model,cna_test,idpc_test,parliament_test,ukusnews_test,mediacorp_test,idpc_short_test,parliament_short_test,ukusnews_short_test,mediacorp_short_test,ytb_asr_batch1,ytb_asr_batch2,seame_dev_man,seame_dev_sge
2
  Qwen-Audio-Chat,0.19753284203780838,0.7710863986313088,0.26279685873781816,0.3158631121194933,0.4498529892192094,0.6008025988916491,0.09347360821020603,0.10399586086125925,0.2548909377108163,0.2297764461857571,0.4315277327278625,0.8783373786407767,1.05567969634822
3
- MERaLiON-AudioLLM-Whisper-SEA-LION,0.15924383210509452,0.30008554319931563,0.058922319992430694,0.12554358101720553,0.170859196341065,0.24918784635964075,0.056935097083623425,0.10144869855926132,0.13301101866426804,0.12222478297610817,0.25844956211482323,0.388282092772384,0.35550521901496834
4
  hy_whisper_local_cs,0.15710776460536152,0.19863130881094954,0.058638471000094616,0.07199848742673473,0.13124795818360013,0.17638066118861073,0.06559913359634872,0.07828544137546764,0.1154711041151338,0.11546439271721595,0.22990593577684074,0.3134101941747573,0.33199669411368576
5
  Qwen2-Audio-7B-Instruct,0.2067713339741536,0.19093242087254064,0.23270886555019396,0.13843826810361126,0.18694870957203527,0.21326199120963119,0.08416492612361723,0.1194380323171217,0.17180121430177647,0.16843358684796805,0.2080008649583739,0.5522518878101402,0.5486546879304539
6
  whisper_large_v3,0.13841717398269784,0.19880239520958085,0.0753619074652285,0.07135564378899603,0.12054884024828487,0.1662526275558953,0.05543951935226013,0.06168908700151238,0.11715763436024286,0.12226319428439733,0.17210509244242622,0.7225930420711975,0.5377268970583734
 
7
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.15171419416853574,0.16766467065868262,0.06282524363705176,0.07388920400831915,0.12455080039202875,0.16931014714313014,0.07325752301384698,0.06877338215394412,0.14571621317742298,0.1400092187139894,0.2192622950819672,0.7824973031283711,0.5840399155162387
8
  gemini-1.5-flash,,,,,,,,,,0.1089344703080587,,0.9690871089536138,1.1100431601824359
9
- WavLLM_fairseq,0.29856883477517887,0.7686911890504705,0.5216434856656259,0.5911892607298166,0.3595230316889905,0.36728454041658704,0.09512390087929656,0.2066783411605508,0.2621992354396222,0.41876008296842593,0.48091685587631094,1.2913969795037756,1.2204842511249197
10
  SALMONN_7B,0.15395706504325538,0.4550898203592814,0.3010928186204939,0.18918510115333712,0.32089186540346293,0.26313777947639977,0.08676929424202573,0.09042426172092653,0.1751742747919946,0.21487285856956287,0.3238620391393664,1.2721817691477886,1.0189782362484312
11
  cascade_whisper_large_v3_llama_3_8b_instruct,0.13798996048275125,0.17741659538066723,0.07517267480367111,0.07642276422764227,0.13598497223129696,0.15803554366520162,0.05742502771975968,0.0700867627159118,0.11434675061839443,0.12579703464700007,0.23561466104443723,0.6848705501618123,0.507882090054792
 
1
  Model,cna_test,idpc_test,parliament_test,ukusnews_test,mediacorp_test,idpc_short_test,parliament_short_test,ukusnews_short_test,mediacorp_short_test,ytb_asr_batch1,ytb_asr_batch2,seame_dev_man,seame_dev_sge
2
  Qwen-Audio-Chat,0.19753284203780838,0.7710863986313088,0.26279685873781816,0.3158631121194933,0.4498529892192094,0.6008025988916491,0.09347360821020603,0.10399586086125925,0.2548909377108163,0.2297764461857571,0.4315277327278625,0.8783373786407767,1.05567969634822
3
+ MERaLiON-AudioLLM-Whisper-SEA-LION,0.15924383210509452,0.30008554319931563,0.058922319992430694,0.12554358101720553,0.170859196341065,0.24918784635964075,0.056935097083623425,0.10144869855926132,0.13301101866426804,0.11484981178458939,0.15162720294085846,0.388282092772384,0.35550521901496834
4
  hy_whisper_local_cs,0.15710776460536152,0.19863130881094954,0.058638471000094616,0.07199848742673473,0.13124795818360013,0.17638066118861073,0.06559913359634872,0.07828544137546764,0.1154711041151338,0.11546439271721595,0.22990593577684074,0.3134101941747573,0.33199669411368576
5
  Qwen2-Audio-7B-Instruct,0.2067713339741536,0.19093242087254064,0.23270886555019396,0.13843826810361126,0.18694870957203527,0.21326199120963119,0.08416492612361723,0.1194380323171217,0.17180121430177647,0.16843358684796805,0.2080008649583739,0.5522518878101402,0.5486546879304539
6
  whisper_large_v3,0.13841717398269784,0.19880239520958085,0.0753619074652285,0.07135564378899603,0.12054884024828487,0.1662526275558953,0.05543951935226013,0.06168908700151238,0.11715763436024286,0.12226319428439733,0.17210509244242622,0.7225930420711975,0.5377268970583734
7
+ old_models,,,,,,,,,,,,,
8
  cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct,0.15171419416853574,0.16766467065868262,0.06282524363705176,0.07388920400831915,0.12455080039202875,0.16931014714313014,0.07325752301384698,0.06877338215394412,0.14571621317742298,0.1400092187139894,0.2192622950819672,0.7824973031283711,0.5840399155162387
9
  gemini-1.5-flash,,,,,,,,,,0.1089344703080587,,0.9690871089536138,1.1100431601824359
10
+ WavLLM_fairseq,0.26946491509131687,0.7686911890504705,0.5216434856656259,0.5911892607298166,0.3595230316889905,0.36728454041658704,0.09512390087929656,0.2066783411605508,0.2621992354396222,0.41876008296842593,0.48091685587631094,1.2913969795037756,1.2204842511249197
11
  SALMONN_7B,0.15395706504325538,0.4550898203592814,0.3010928186204939,0.18918510115333712,0.32089186540346293,0.26313777947639977,0.08676929424202573,0.09042426172092653,0.1751742747919946,0.21487285856956287,0.3238620391393664,1.2721817691477886,1.0189782362484312
12
  cascade_whisper_large_v3_llama_3_8b_instruct,0.13798996048275125,0.17741659538066723,0.07517267480367111,0.07642276422764227,0.13598497223129696,0.15803554366520162,0.05742502771975968,0.0700867627159118,0.11434675061839443,0.12579703464700007,0.23561466104443723,0.6848705501618123,0.507882090054792