Hugging Face's logo

Spaces:
MERaLiON
/
AudioBench-Leaderboard
Running

App Files Community
zhuohan-7 commited on
Commit
ec6e789
•
1 Parent(s): cde78b8

Upload folder using huggingface_hub

Browse files
Files changed (12) hide show
  1. results/bleu/st.csv +0 -4
  2. results/llama3_70b_judge/ac.csv +0 -4
  3. results/llama3_70b_judge/aqa.csv +0 -4
  4. results/llama3_70b_judge/ar.csv +0 -4
  5. results/llama3_70b_judge/si.csv +0 -4
  6. results/llama3_70b_judge/sqa.csv +0 -4
  7. results/llama3_70b_judge_binary/er.csv +0 -4
  8. results/llama3_70b_judge_binary/gr.csv +0 -4
  9. results/llama3_70b_judge_binary/sqa.csv +0 -4
  10. results/meteor/ac.csv +0 -4
  11. results/wer/asr.csv +0 -4
  12. results/wer/cnasr.csv +0 -4
results/bleu/st.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,covost2_en_id_test,covost2_en_zh_test,covost2_en_ta_test,covost2_id_en_test,covost2_zh_en_test,covost2_ta_en_test
2
- AudioGemma2_v1,,,,,,
3
- test_temp,,,,,,
4
- audiogemma_2_singlish,,,,,,
5
  salmonn_7b,14.135708265627702,34.055681569415356,0.00045042243733421275,27.877196356747618,5.2319724257915645,0.4256092994569619
6
  wavllm_fairseq,13.841886973016162,31.96381187282953,0.0033159224040994286,5.933522277713613,2.368659001743569,0.1695522548322915
7
- old_models,,,,,,
8
  Qwen2-Audio-7B-Instruct,16.325186897428104,25.765420247070075,0.03245972071872916,6.326113431899141,16.466557744958333,0.04425838146050298
9
  whisper_large_v3_with_llama_3_8b_instruct,,,,,,
10
  mowe_audio,,,,,,
 
1
  Model,covost2_en_id_test,covost2_en_zh_test,covost2_en_ta_test,covost2_id_en_test,covost2_zh_en_test,covost2_ta_en_test
 
 
 
2
  salmonn_7b,14.135708265627702,34.055681569415356,0.00045042243733421275,27.877196356747618,5.2319724257915645,0.4256092994569619
3
  wavllm_fairseq,13.841886973016162,31.96381187282953,0.0033159224040994286,5.933522277713613,2.368659001743569,0.1695522548322915
 
4
  Qwen2-Audio-7B-Instruct,16.325186897428104,25.765420247070075,0.03245972071872916,6.326113431899141,16.466557744958333,0.04425838146050298
5
  whisper_large_v3_with_llama_3_8b_instruct,,,,,,
6
  mowe_audio,,,,,,
results/llama3_70b_judge/ac.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,audiocaps_test,wavcaps_test
2
- AudioGemma2_v1,,
3
- test_temp,,
4
- audiogemma_2_singlish,,
5
  salmonn_7b,34.372727272727275,21.15606936416185
6
  wavllm_fairseq,5.5,6.901734104046243
7
- old_models,,
8
  Qwen2-Audio-7B-Instruct,40.77727272727273,33.78034682080925
9
  whisper_large_v3_with_llama_3_8b_instruct,2.4727272727272727,3.445086705202312
10
  mowe_audio,41.20909090909091,23.294797687861273
 
1
  Model,audiocaps_test,wavcaps_test
 
 
 
2
  salmonn_7b,34.372727272727275,21.15606936416185
3
  wavllm_fairseq,5.5,6.901734104046243
 
4
  Qwen2-Audio-7B-Instruct,40.77727272727273,33.78034682080925
5
  whisper_large_v3_with_llama_3_8b_instruct,2.4727272727272727,3.445086705202312
6
  mowe_audio,41.20909090909091,23.294797687861273
results/llama3_70b_judge/aqa.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,clotho_aqa_test,audiocaps_qa_test,wavcaps_qa_test
2
- AudioGemma2_v1,,,
3
- test_temp,,,
4
- audiogemma_2_singlish,,,
5
  salmonn_7b,55.75299866725899,47.02875399361022,46.25
6
  wavllm_fairseq,43.01199466903598,29.840255591054312,26.25
7
- old_models,,,
8
  Qwen2-Audio-7B-Instruct,50.919591292758774,45.75079872204473,44.473684210526315
9
  whisper_large_v3_with_llama_3_8b_instruct,29.47134606841404,17.380191693290733,16.710526315789473
10
  mowe_audio,62.221235006663704,32.97124600638978,28.88157894736842
 
1
  Model,clotho_aqa_test,audiocaps_qa_test,wavcaps_qa_test
 
 
 
2
  salmonn_7b,55.75299866725899,47.02875399361022,46.25
3
  wavllm_fairseq,43.01199466903598,29.840255591054312,26.25
 
4
  Qwen2-Audio-7B-Instruct,50.919591292758774,45.75079872204473,44.473684210526315
5
  whisper_large_v3_with_llama_3_8b_instruct,29.47134606841404,17.380191693290733,16.710526315789473
6
  mowe_audio,62.221235006663704,32.97124600638978,28.88157894736842
results/llama3_70b_judge/ar.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,voxceleb_accent_test
2
- AudioGemma2_v1,
3
- test_temp,
4
- audiogemma_2_singlish,
5
  salmonn_7b,37.443578169881
6
  wavllm_fairseq,
7
- old_models,
8
  Qwen2-Audio-7B-Instruct,29.187525646286417
9
  whisper_large_v3_with_llama_3_8b_instruct,39.32704144439885
10
  mowe_audio,23.68485843249897
 
1
  Model,voxceleb_accent_test
 
 
 
2
  salmonn_7b,37.443578169881
3
  wavllm_fairseq,
 
4
  Qwen2-Audio-7B-Instruct,29.187525646286417
5
  whisper_large_v3_with_llama_3_8b_instruct,39.32704144439885
6
  mowe_audio,23.68485843249897
results/llama3_70b_judge/si.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,openhermes_audio_test,alpaca_audio_test
2
- AudioGemma2_v1,,
3
- test_temp,,
4
- audiogemma_2_singlish,,
5
  salmonn_7b,19.2,12.4
6
  wavllm_fairseq,22.400000000000002,21.6
7
- old_models,,
8
  Qwen2-Audio-7B-Instruct,44.800000000000004,52.599999999999994
9
  whisper_large_v3_with_llama_3_8b_instruct,63.0,70.8
10
  mowe_audio,16.0,19.8
 
1
  Model,openhermes_audio_test,alpaca_audio_test
 
 
 
2
  salmonn_7b,19.2,12.4
3
  wavllm_fairseq,22.400000000000002,21.6
 
4
  Qwen2-Audio-7B-Instruct,44.800000000000004,52.599999999999994
5
  whisper_large_v3_with_llama_3_8b_instruct,63.0,70.8
6
  mowe_audio,16.0,19.8
results/llama3_70b_judge/sqa.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,slue_p2_sqa5_test,public_sg_speech_qa_test,spoken_squad_v1
2
- AudioGemma2_v1,,,
3
- test_temp,,,
4
- audiogemma_2_singlish,,,
5
  salmonn_7b,78.23529411764706,56.77325581395348,
6
  wavllm_fairseq,83.92156862745098,58.54651162790698,
7
- old_models,,,
8
  Qwen2-Audio-7B-Instruct,80.04901960784315,58.31395348837209,
9
  whisper_large_v3_with_llama_3_8b_instruct,82.99019607843137,64.94186046511628,
10
  mowe_audio,76.5686274509804,,
 
1
  Model,slue_p2_sqa5_test,public_sg_speech_qa_test,spoken_squad_v1
 
 
 
2
  salmonn_7b,78.23529411764706,56.77325581395348,
3
  wavllm_fairseq,83.92156862745098,58.54651162790698,
 
4
  Qwen2-Audio-7B-Instruct,80.04901960784315,58.31395348837209,
5
  whisper_large_v3_with_llama_3_8b_instruct,82.99019607843137,64.94186046511628,
6
  mowe_audio,76.5686274509804,,
results/llama3_70b_judge_binary/er.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,iemocap_emotion_test,meld_sentiment_test,meld_emotion_test
2
- AudioGemma2_v1,,,
3
- test_temp,,,
4
- audiogemma_2_singlish,,,
5
  salmonn_7b,21.55688622754491,41.877394636015325,33.06513409961686
6
  wavllm_fairseq,45.90818363273453,50.076628352490424,41.0727969348659
7
- old_models,,,
8
  Qwen2-Audio-7B-Instruct,49.30139720558882,53.486590038314176,40.536398467432946
9
  whisper_large_v3_with_llama_3_8b_instruct,34.4311377245509,43.86973180076628,33.25670498084291
10
  mowe_audio,23.55289421157685,46.09195402298851,30.07662835249042
 
1
  Model,iemocap_emotion_test,meld_sentiment_test,meld_emotion_test
 
 
 
2
  salmonn_7b,21.55688622754491,41.877394636015325,33.06513409961686
3
  wavllm_fairseq,45.90818363273453,50.076628352490424,41.0727969348659
 
4
  Qwen2-Audio-7B-Instruct,49.30139720558882,53.486590038314176,40.536398467432946
5
  whisper_large_v3_with_llama_3_8b_instruct,34.4311377245509,43.86973180076628,33.25670498084291
6
  mowe_audio,23.55289421157685,46.09195402298851,30.07662835249042
results/llama3_70b_judge_binary/gr.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,voxceleb_gender_test,iemocap_gender_test
2
- AudioGemma2_v1,,
3
- test_temp,,
4
- audiogemma_2_singlish,,
5
  salmonn_7b,88.90028723840788,51.59680638722555
6
  wavllm_fairseq,69.67583093967994,
7
- old_models,,
8
  Qwen2-Audio-7B-Instruct,99.1177677472302,49.30139720558882
9
  whisper_large_v3_with_llama_3_8b_instruct,53.40582683627411,51.49700598802395
10
  mowe_audio,81.20640131308986,37.82435129740519
 
1
  Model,voxceleb_gender_test,iemocap_gender_test
 
 
 
2
  salmonn_7b,88.90028723840788,51.59680638722555
3
  wavllm_fairseq,69.67583093967994,
 
4
  Qwen2-Audio-7B-Instruct,99.1177677472302,49.30139720558882
5
  whisper_large_v3_with_llama_3_8b_instruct,53.40582683627411,51.49700598802395
6
  mowe_audio,81.20640131308986,37.82435129740519
results/llama3_70b_judge_binary/sqa.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,cn_college_listen_mcq_test,dream_tts_mcq_test
2
- AudioGemma2_v1,,
3
- test_temp,,
4
- audiogemma_2_singlish,,
5
  salmonn_7b,50.5063848524879,55.933089388395196
6
  wavllm_fairseq,65.43372963452224,64.55828541557763
7
- old_models,,
8
  Qwen2-Audio-7B-Instruct,74.50462351387054,66.70151594354418
9
  whisper_large_v3_with_llama_3_8b_instruct,85.24878907970057,86.0951385258756
10
  mowe_audio,75.3412593571114,
 
1
  Model,cn_college_listen_mcq_test,dream_tts_mcq_test
 
 
 
2
  salmonn_7b,50.5063848524879,55.933089388395196
3
  wavllm_fairseq,65.43372963452224,64.55828541557763
 
4
  Qwen2-Audio-7B-Instruct,74.50462351387054,66.70151594354418
5
  whisper_large_v3_with_llama_3_8b_instruct,85.24878907970057,86.0951385258756
6
  mowe_audio,75.3412593571114,
results/meteor/ac.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,audiocaps_test,wavcaps_test
2
- AudioGemma2_v1,,
3
- test_temp,,
4
- audiogemma_2_singlish,,
5
  salmonn_7b,,
6
  wavllm_fairseq,,
7
- old_models,,
8
  Qwen2-Audio-7B-Instruct,,
9
  whisper_large_v3_with_llama_3_8b_instruct,,
10
  mowe_audio,,
 
1
  Model,audiocaps_test,wavcaps_test
 
 
 
2
  salmonn_7b,,
3
  wavllm_fairseq,,
 
4
  Qwen2-Audio-7B-Instruct,,
5
  whisper_large_v3_with_llama_3_8b_instruct,,
6
  mowe_audio,,
results/wer/asr.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,librispeech_test_clean,librispeech_test_other,common_voice_15_en_test,peoples_speech_test,gigaspeech_test,earnings21_test,earnings22_test,tedlium3_test,tedlium3_long_form_test,imda_part1_asr_test,imda_part2_asr_test
2
- AudioGemma2_v1,0.03089749877390878,0.06278018423591276,0.10836691280904365,0.2604876551258585,0.10853395305460918,0.16500928283176397,,,,,
3
- test_temp,,,,,,,,,,,
4
- audiogemma_2_singlish,0.026653336854415816,0.05710556679970492,0.10400318764018227,0.2569853056468371,0.10561083409168967,0.1646007270625118,,,,,
5
  salmonn_7b,0.555834307918663,0.4180490665254317,0.3375363911542985,0.34335568653146914,0.14226764839341272,0.2687240831119165,0.3638199574803293,0.08558958163301515,0.18386810849140223,0.1581255622670608,0.6915082052617869
6
  wavllm_fairseq,0.02103218017882069,0.04798834811886432,0.14533325621300636,0.3792176325635977,0.15491778414546403,0.6447482518259942,0.6671766188447099,0.06621482559171073,0.4536784258110264,0.07762498393522684,0.4511070591299818
7
- old_models,,,,,,,,,,,
8
  Qwen2-Audio-7B-Instruct,0.03201041234390915,0.06073732195887794,0.11438872500819404,0.22323599266520214,0.11891779417331638,0.9865371146513101,0.9884312836905533,0.06387282211419042,0.953518879631271,,
9
  whisper_large_v3_with_llama_3_8b_instruct,0.01831591655034519,0.03714982881570734,0.09876543209876543,0.14540692118393275,0.09515429104337297,0.11773910240019567,0.15611126487402763,0.038146268762641496,0.04754476156709803,0.0464432592211798,0.2904141703568638
10
  mowe_audio,0.023258007318821442,0.05544101235175062,0.5609539784448686,0.6840078629561348,0.32259099649275863,0.32744494224633414,0.41765995821028074,0.2092544622263227,0.16780712639602907,,
 
1
  Model,librispeech_test_clean,librispeech_test_other,common_voice_15_en_test,peoples_speech_test,gigaspeech_test,earnings21_test,earnings22_test,tedlium3_test,tedlium3_long_form_test,imda_part1_asr_test,imda_part2_asr_test
 
 
 
2
  salmonn_7b,0.555834307918663,0.4180490665254317,0.3375363911542985,0.34335568653146914,0.14226764839341272,0.2687240831119165,0.3638199574803293,0.08558958163301515,0.18386810849140223,0.1581255622670608,0.6915082052617869
3
  wavllm_fairseq,0.02103218017882069,0.04798834811886432,0.14533325621300636,0.3792176325635977,0.15491778414546403,0.6447482518259942,0.6671766188447099,0.06621482559171073,0.4536784258110264,0.07762498393522684,0.4511070591299818
 
4
  Qwen2-Audio-7B-Instruct,0.03201041234390915,0.06073732195887794,0.11438872500819404,0.22323599266520214,0.11891779417331638,0.9865371146513101,0.9884312836905533,0.06387282211419042,0.953518879631271,,
5
  whisper_large_v3_with_llama_3_8b_instruct,0.01831591655034519,0.03714982881570734,0.09876543209876543,0.14540692118393275,0.09515429104337297,0.11773910240019567,0.15611126487402763,0.038146268762641496,0.04754476156709803,0.0464432592211798,0.2904141703568638
6
  mowe_audio,0.023258007318821442,0.05544101235175062,0.5609539784448686,0.6840078629561348,0.32259099649275863,0.32744494224633414,0.41765995821028074,0.2092544622263227,0.16780712639602907,,
results/wer/cnasr.csv CHANGED
@@ -1,10 +1,6 @@
1
  Model,aishell_asr_zh_test
2
- AudioGemma2_v1,
3
- test_temp,
4
- audiogemma_2_singlish,
5
  salmonn_7b,0.8527647443131199
6
  wavllm_fairseq,0.7054601967888183
7
- old_models,
8
  Qwen2-Audio-7B-Instruct,0.09317772366415236
9
  whisper_large_v3_with_llama_3_8b_instruct,
10
  mowe_audio,1.0174318465284788
 
1
  Model,aishell_asr_zh_test
 
 
 
2
  salmonn_7b,0.8527647443131199
3
  wavllm_fairseq,0.7054601967888183
 
4
  Qwen2-Audio-7B-Instruct,0.09317772366415236
5
  whisper_large_v3_with_llama_3_8b_instruct,
6
  mowe_audio,1.0174318465284788