Upload organize_model_results.json with huggingface_hub

organize_model_results.json

@@ -5,6 +5,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 47.01682396389003,
             "Qwen2-Audio-7B-Instruct": 29.187525646286417,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 24.640951990151827,
+            "phi_4_multimodal_instruct": 26.815757078375054,
             "WavLLM_fairseq": 39.96717275338531,
             "SALMONN_7B": 34.222404595814524,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 39.32704144439885
@@ -13,12 +14,22 @@
             "cascade_whisper_large_v3_llama_3_8b_instruct": 39.462453836684446
         }
     },
+    "imda_part4_30s_asr_test": {
+        "wer": {
+            "Qwen-Audio-Chat": 1.1764312018747907,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.0,
+            "Qwen2-Audio-7B-Instruct": 0.5685405990059489,
+            "whisper_large_v3": 0.8294532718704128,
+            "phi_4_multimodal_instruct": 1.3868687388941825
+        }
+    },
     "wavcaps_test": {
         "meteor": {
             "Qwen-Audio-Chat": 0.2355106805560457,
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.3175511907248581,
             "Qwen2-Audio-7B-Instruct": 0.21342294856199182,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.120421856260385,
+            "phi_4_multimodal_instruct": 0.24508284335582894,
             "WavLLM_fairseq": 0.06399522524688675,
             "SALMONN_7B": 0.17175112770658157,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.1388630786594543
@@ -28,6 +39,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 33.97687861271676,
             "Qwen2-Audio-7B-Instruct": 33.78034682080925,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 6.3468208092485545,
+            "phi_4_multimodal_instruct": 21.884393063583815,
             "WavLLM_fairseq": 6.901734104046243,
             "SALMONN_7B": 23.76878612716763,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 3.445086705202312
@@ -43,6 +55,7 @@
             "Qwen2-Audio-7B-Instruct": 16.466557744958333,
             "whisper_large_v3": 14.673689493155793,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 15.209998552437538,
+            "phi_4_multimodal_instruct": 22.678131781242936,
             "WavLLM_fairseq": 2.368659001743569,
             "SALMONN_7B": 5.296039450108202,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 14.154700735606419
@@ -54,6 +67,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 67.2,
             "Qwen2-Audio-7B-Instruct": 53.6,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 71.6,
+            "phi_4_multimodal_instruct": 66.2,
             "WavLLM_fairseq": 62.199999999999996,
             "SALMONN_7B": 46.8,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 64.0
@@ -92,6 +106,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 57.7927548441449,
             "Qwen2-Audio-7B-Instruct": 71.60909856781802,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 51.727042965459134,
+            "phi_4_multimodal_instruct": 54.422914911541696,
             "WavLLM_fairseq": 44.3133951137321,
             "SALMONN_7B": 50.88458298230834,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 56.44481887110362
@@ -116,6 +131,7 @@
             "Qwen2-Audio-7B-Instruct": 0.1905689473257041,
             "whisper_large_v3": 0.3171008846684522,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.32988393799204613,
+            "phi_4_multimodal_instruct": 0.3470091713334957,
             "WavLLM_fairseq": 0.4463923382842302,
             "SALMONN_7B": 0.42346400454508565,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.31912994075156237
@@ -128,6 +144,7 @@
             "Qwen2-Audio-7B-Instruct": 0.18872219319407232,
             "whisper_large_v3": 0.11863959266711877,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.11416493424197618,
+            "phi_4_multimodal_instruct": 0.15921168191570967,
             "WavLLM_fairseq": 0.6447482518259942,
             "SALMONN_7B": 0.2577708974886327,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.11773910240019567
@@ -152,6 +169,7 @@
             "Qwen2-Audio-7B-Instruct": 0.060415760304159495,
             "whisper_large_v3": 0.03660128246354058,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.05307658841999735,
+            "phi_4_multimodal_instruct": 0.03879546787220762,
             "WavLLM_fairseq": 0.04798834811886432,
             "SALMONN_7B": 0.09671439650443565,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.03714982881570734
@@ -164,6 +182,7 @@
             "Qwen2-Audio-7B-Instruct": 0.035141660693401744,
             "whisper_large_v3": 0.01878749009695552,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.032349945297468596,
+            "phi_4_multimodal_instruct": 0.0167502923755989,
             "WavLLM_fairseq": 0.02103218017882069,
             "SALMONN_7B": 0.10270871845172973,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.018334779492209605
@@ -176,6 +195,7 @@
             "Qwen2-Audio-7B-Instruct": 0.2245352799625317,
             "whisper_large_v3": 0.1698509342851144,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.1789273082575623,
+            "phi_4_multimodal_instruct": 0.14552883606001388,
             "WavLLM_fairseq": 0.42541061709652933,
             "SALMONN_7B": 0.24872817713464365,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.17467982364056267
@@ -187,6 +207,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 65.6,
             "Qwen2-Audio-7B-Instruct": 44.800000000000004,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 72.2,
+            "phi_4_multimodal_instruct": 30.8,
             "WavLLM_fairseq": 19.2,
             "SALMONN_7B": 15.8,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 63.0
@@ -201,6 +222,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 48.505976095617534,
             "Qwen2-Audio-7B-Instruct": 53.98406374501992,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 44.322709163346616,
+            "phi_4_multimodal_instruct": 41.03585657370518,
             "WavLLM_fairseq": 59.76095617529881,
             "SALMONN_7B": 23.804780876494025,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 46.713147410358566
@@ -212,6 +234,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 59.7093023255814,
             "Qwen2-Audio-7B-Instruct": 58.31395348837209,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 73.11046511627907,
+            "phi_4_multimodal_instruct": 68.40116279069767,
             "WavLLM_fairseq": 58.54651162790698,
             "SALMONN_7B": 59.24418604651163,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 64.94186046511628
@@ -239,6 +262,7 @@
             "Qwen2-Audio-7B-Instruct": 0.11438872500819404,
             "whisper_large_v3": 0.10001863741235596,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.10600831614192711,
+            "phi_4_multimodal_instruct": 0.08262800367606891,
             "WavLLM_fairseq": 0.14533325621300636,
             "SALMONN_7B": 0.3062255383962828,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.09876543209876543
@@ -250,6 +274,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 73.66473556344609,
             "Qwen2-Audio-7B-Instruct": 64.86264249672958,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 88.61894972902262,
+            "phi_4_multimodal_instruct": 77.58549803774996,
             "WavLLM_fairseq": 77.64903756307233,
             "SALMONN_7B": 66.39506634273968,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 83.81984675761541
@@ -266,6 +291,7 @@
             "whisper_large_v3": 0.5377268970583734,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.5840399155162387,
             "gemini-1.5-flash": 1.1100431601824359,
+            "phi_4_multimodal_instruct": 0.8529492791331231,
             "WavLLM_fairseq": 1.2204842511249197,
             "SALMONN_7B": 1.0189782362484312,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.507882090054792
@@ -277,6 +303,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 46.206896551724135,
             "Qwen2-Audio-7B-Instruct": 53.9463601532567,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 56.59003831417625,
+            "phi_4_multimodal_instruct": 51.609195402298845,
             "WavLLM_fairseq": 51.072796934865906,
             "SALMONN_7B": 41.7624521072797,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 45.593869731800766
@@ -288,6 +315,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 53.2,
             "Qwen2-Audio-7B-Instruct": 39.6,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 66.0,
+            "phi_4_multimodal_instruct": 43.8,
             "WavLLM_fairseq": 46.6,
             "SALMONN_7B": 36.6,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 53.8
@@ -302,6 +330,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 99.75379565038982,
             "Qwen2-Audio-7B-Instruct": 99.1177677472302,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 34.94050061551087,
+            "phi_4_multimodal_instruct": 94.54247025030776,
             "WavLLM_fairseq": 69.61427985227739,
             "SALMONN_7B": 88.79770209273697,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 42.921624948707425
@@ -313,6 +342,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 93.76666666666667,
             "Qwen2-Audio-7B-Instruct": 61.56666666666667,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 19.6,
+            "phi_4_multimodal_instruct": 36.833333333333336,
             "WavLLM_fairseq": 46.766666666666666,
             "SALMONN_7B": 42.733333333333334,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 25.433333333333337
@@ -337,6 +367,7 @@
             "Qwen2-Audio-7B-Instruct": 0.27856006770658537,
             "whisper_large_v3": 0.2143555471246589,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.22881615619208825,
+            "phi_4_multimodal_instruct": 0.22801359968481416,
             "WavLLM_fairseq": 0.39796588405247263,
             "SALMONN_7B": 0.34868891450584405,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.22004640235805695
@@ -360,6 +391,7 @@
             "Qwen2-Audio-7B-Instruct": 0.23542555661330924,
             "whisper_large_v3": 0.15887899737116104,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.1448629161356777,
+            "phi_4_multimodal_instruct": 0.24134627375003423,
             "WavLLM_fairseq": 0.6671766188447099,
             "SALMONN_7B": 0.3597423676988383,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.15611126487402763
@@ -396,6 +428,7 @@
             "Qwen2-Audio-7B-Instruct": 6.326113431899141,
             "whisper_large_v3": 46.01512198258627,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 46.80524126004861,
+            "phi_4_multimodal_instruct": 0.36465303013961253,
             "WavLLM_fairseq": 5.933522277713613,
             "SALMONN_7B": 26.89649039333571,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 46.79924664837527
@@ -432,6 +465,7 @@
             "Qwen2-Audio-7B-Instruct": 0.11723812890302816,
             "whisper_large_v3": 0.09459022434812692,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.09948381629977261,
+            "phi_4_multimodal_instruct": 0.09672866386388193,
             "WavLLM_fairseq": 0.15491778414546403,
             "SALMONN_7B": 0.10765150204693537,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.09515429104337297
@@ -455,6 +489,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 48.4,
             "Qwen2-Audio-7B-Instruct": 33.8,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 45.4,
+            "phi_4_multimodal_instruct": 41.2,
             "WavLLM_fairseq": 31.6,
             "SALMONN_7B": 9.0,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 37.400000000000006
@@ -469,6 +504,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 77.83333333333333,
             "Qwen2-Audio-7B-Instruct": 0.9666666666666667,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 7.633333333333334,
+            "phi_4_multimodal_instruct": 0.5333333333333333,
             "WavLLM_fairseq": 0.23333333333333336,
             "SALMONN_7B": 0.06666666666666667,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 9.666666666666666
@@ -480,6 +516,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 93.48605577689243,
             "Qwen2-Audio-7B-Instruct": 92.80876494023903,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 15.737051792828685,
+            "phi_4_multimodal_instruct": 59.46215139442231,
             "WavLLM_fairseq": 51.932270916334666,
             "SALMONN_7B": 81.31474103585658,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 44.22310756972111
@@ -492,6 +529,7 @@
             "Qwen2-Audio-7B-Instruct": 0.2080008649583739,
             "whisper_large_v3": 0.17210509244242622,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.2192622950819672,
+            "phi_4_multimodal_instruct": 0.23849064763758243,
             "WavLLM_fairseq": 0.48091685587631094,
             "SALMONN_7B": 0.3238620391393664,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.23561466104443723
@@ -516,6 +554,7 @@
             "Qwen2-Audio-7B-Instruct": 74.7247908410392,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 91.85380889476001,
             "gemini-1.5-flash": 89.25583443416997,
+            "phi_4_multimodal_instruct": 73.18361955085865,
             "WavLLM_fairseq": 66.31439894319684,
             "SALMONN_7B": 50.99075297225891,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 85.2928225451343
@@ -527,6 +566,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 84.31782540512285,
             "Qwen2-Audio-7B-Instruct": 66.49242028227914,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 89.33612127548353,
+            "phi_4_multimodal_instruct": 72.60846837428123,
             "WavLLM_fairseq": 66.5446941975954,
             "SALMONN_7B": 56.455828541557764,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 86.4610559330894
@@ -538,6 +578,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 57.0,
             "Qwen2-Audio-7B-Instruct": 40.4,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 58.0,
+            "phi_4_multimodal_instruct": 52.199999999999996,
             "WavLLM_fairseq": 45.199999999999996,
             "SALMONN_7B": 17.2,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 49.0
@@ -553,6 +594,7 @@
             "Qwen2-Audio-7B-Instruct": 0.09260359129694522,
             "whisper_large_v3": 0.12359684029221357,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.20886539565639167,
+            "phi_4_multimodal_instruct": 0.07466690423868068,
             "WavLLM_fairseq": 0.7054601967888183,
             "SALMONN_7B": 0.8259290055631446,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.12450753301261111
@@ -564,6 +606,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 51.4,
             "Qwen2-Audio-7B-Instruct": 42.0,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 56.0,
+            "phi_4_multimodal_instruct": 43.8,
             "WavLLM_fairseq": 45.199999999999996,
             "SALMONN_7B": 40.599999999999994,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 49.0
@@ -578,6 +621,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 46.4,
             "Qwen2-Audio-7B-Instruct": 24.8,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 44.0,
+            "phi_4_multimodal_instruct": 37.0,
             "WavLLM_fairseq": 31.6,
             "SALMONN_7B": 7.0,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 36.0
@@ -604,6 +648,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 38.00454545454545,
             "Qwen2-Audio-7B-Instruct": 40.77727272727273,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 3.0954545454545457,
+            "phi_4_multimodal_instruct": 26.386363636363637,
             "WavLLM_fairseq": 5.5,
             "SALMONN_7B": 37.445454545454545,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 2.4727272727272727
@@ -613,6 +658,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.24920047034353812,
             "Qwen2-Audio-7B-Instruct": 0.19891712076314283,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.05796819723943051,
+            "phi_4_multimodal_instruct": 0.1757379026471828,
             "WavLLM_fairseq": 0.041732965094428545,
             "SALMONN_7B": 0.20994052484339956,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.07953048457785493
@@ -627,6 +673,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 7.816666666666666,
             "Qwen2-Audio-7B-Instruct": 2.55,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 26.016666666666666,
+            "phi_4_multimodal_instruct": 3.5166666666666666,
             "WavLLM_fairseq": 2.6833333333333336,
             "SALMONN_7B": 2.5166666666666666,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 12.416666666666666
@@ -651,6 +698,7 @@
             "Qwen2-Audio-7B-Instruct": 0.04425838146050298,
             "whisper_large_v3": 2.451098639578599,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 2.8327095799289337,
+            "phi_4_multimodal_instruct": 0.053138495633157125,
             "WavLLM_fairseq": 0.1695522548322915,
             "SALMONN_7B": 0.3649023706010388,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 2.4245628096245917
@@ -663,6 +711,7 @@
             "Qwen2-Audio-7B-Instruct": 16.325186897428104,
             "whisper_large_v3": 1.600581653970121,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 27.620150160643625,
+            "phi_4_multimodal_instruct": 15.012558278964478,
             "WavLLM_fairseq": 13.841886973016162,
             "SALMONN_7B": 14.102682915273142,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 10.930203684508578
@@ -674,6 +723,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 63.15021876519203,
             "Qwen2-Audio-7B-Instruct": 50.919591292758774,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 24.647544968400585,
+            "phi_4_multimodal_instruct": 47.86582401555663,
             "WavLLM_fairseq": 43.01199466903598,
             "SALMONN_7B": 57.75401069518716,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 29.47134606841404
@@ -690,6 +740,7 @@
             "whisper_large_v3": 0.12226319428439733,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.1400092187139894,
             "gemini-1.5-flash": 0.1089344703080587,
+            "phi_4_multimodal_instruct": 0.16175001920565416,
             "WavLLM_fairseq": 0.41876008296842593,
             "SALMONN_7B": 0.21487285856956287,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.12579703464700007
@@ -702,6 +753,7 @@
             "Qwen2-Audio-7B-Instruct": 0.35076166942732234,
             "whisper_large_v3": 0.27026366524560785,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.3035544573275043,
+            "phi_4_multimodal_instruct": 0.44227061666711925,
             "WavLLM_fairseq": 0.7540934640345399,
             "SALMONN_7B": 0.6569229098215983,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.29992939962527493
@@ -713,6 +765,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 74.80000000000001,
             "Qwen2-Audio-7B-Instruct": 52.599999999999994,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 73.8,
+            "phi_4_multimodal_instruct": 25.8,
             "WavLLM_fairseq": 21.6,
             "SALMONN_7B": 17.2,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 70.8
@@ -750,6 +803,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 49.77635782747604,
             "Qwen2-Audio-7B-Instruct": 45.75079872204473,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 18.466453674121407,
+            "phi_4_multimodal_instruct": 38.466453674121404,
             "WavLLM_fairseq": 29.840255591054312,
             "SALMONN_7B": 50.287539936102235,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 17.380191693290733
@@ -765,6 +819,7 @@
             "Qwen2-Audio-7B-Instruct": 0.07197717796796138,
             "whisper_large_v3": 0.06844171360300393,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.07041669714480775,
+            "phi_4_multimodal_instruct": 0.05739643527661961,
             "WavLLM_fairseq": 0.10077292565771828,
             "SALMONN_7B": 0.0925804013361617,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.06922195401458074
@@ -789,6 +844,7 @@
             "Qwen2-Audio-7B-Instruct": 0.03245972071872916,
             "whisper_large_v3": 0.02107778621423822,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 8.433062902024755,
+            "phi_4_multimodal_instruct": 0.19835914151649442,
             "WavLLM_fairseq": 0.0033159224040994286,
             "SALMONN_7B": 0.00046745670226766583,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 1.0368044741318085
@@ -800,6 +856,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 46.31578947368421,
             "Qwen2-Audio-7B-Instruct": 44.473684210526315,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 18.88157894736842,
+            "phi_4_multimodal_instruct": 35.13157894736842,
             "WavLLM_fairseq": 26.25,
             "SALMONN_7B": 47.30263157894737,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 16.710526315789473
@@ -827,11 +884,68 @@
             "Qwen2-Audio-7B-Instruct": 25.765420247070075,
             "whisper_large_v3": 0.16408986541757878,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 35.274306071307024,
+            "phi_4_multimodal_instruct": 45.295964957544776,
             "WavLLM_fairseq": 31.96381187282953,
             "SALMONN_7B": 33.88941292215531,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 5.987143868370054
         }
     },
+    "mmau_mini": {
+        "string_match": {
+            "Qwen-Audio-Chat": 38.5,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 60.5,
+            "Qwen2-Audio-7B-Instruct": 44.4,
+            "phi_4_multimodal_instruct": 54.50000000000001
+        },
+        "llama3_70b_judge": {
+            "Qwen-Audio-Chat": 53.6,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 64.60000000000001,
+            "Qwen2-Audio-7B-Instruct": 58.9,
+            "phi_4_multimodal_instruct": 59.4
+        }
+    },
+    "mmau_mini_music": {
+        "string_match": {
+            "Qwen-Audio-Chat": 0.4311377245508982,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.6077844311377245,
+            "Qwen2-Audio-7B-Instruct": 0.45808383233532934,
+            "phi_4_multimodal_instruct": 0.6377245508982036
+        },
+        "llama3_70b_judge": {
+            "Qwen-Audio-Chat": 0.5958083832335329,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.6437125748502994,
+            "Qwen2-Audio-7B-Instruct": 0.6017964071856288,
+            "phi_4_multimodal_instruct": 0.688622754491018
+        }
+    },
+    "mmau_mini_sound": {
+        "string_match": {
+            "Qwen-Audio-Chat": 0.43543543543543545,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.6606606606606606,
+            "Qwen2-Audio-7B-Instruct": 0.4744744744744745,
+            "phi_4_multimodal_instruct": 0.5975975975975976
+        },
+        "llama3_70b_judge": {
+            "Qwen-Audio-Chat": 0.5945945945945946,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.7027027027027027,
+            "Qwen2-Audio-7B-Instruct": 0.6306306306306306,
+            "phi_4_multimodal_instruct": 0.6456456456456456
+        }
+    },
+    "mmau_mini_speech": {
+        "string_match": {
+            "Qwen-Audio-Chat": 0.2882882882882883,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.5465465465465466,
+            "Qwen2-Audio-7B-Instruct": 0.3993993993993994,
+            "phi_4_multimodal_instruct": 0.3993993993993994
+        },
+        "llama3_70b_judge": {
+            "Qwen-Audio-Chat": 0.4174174174174174,
+            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.5915915915915916,
+            "Qwen2-Audio-7B-Instruct": 0.5345345345345346,
+            "phi_4_multimodal_instruct": 0.44744744744744747
+        }
+    },
     "imda_part5_30s_sqa_test": {
         "llama3_70b_judge": {
             "Qwen-Audio-Chat": 61.260000000000005,
@@ -850,6 +964,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 66.13333333333333,
             "Qwen2-Audio-7B-Instruct": 68.38333333333333,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 26.35,
+            "phi_4_multimodal_instruct": 51.68333333333334,
             "WavLLM_fairseq": 49.06666666666666,
             "SALMONN_7B": 59.766666666666666,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 36.016666666666666
@@ -861,6 +976,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 86.76470588235293,
             "Qwen2-Audio-7B-Instruct": 80.04901960784315,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 88.57843137254902,
+            "phi_4_multimodal_instruct": 88.33333333333334,
             "WavLLM_fairseq": 83.92156862745098,
             "SALMONN_7B": 83.48039215686273,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 82.99019607843137
@@ -876,6 +992,7 @@
             "Qwen2-Audio-7B-Instruct": 0.08739585179932637,
             "whisper_large_v3": 0.03208650948413402,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.04396383619925545,
+            "phi_4_multimodal_instruct": 0.0381847190214501,
             "WavLLM_fairseq": 0.4536784258110264,
             "SALMONN_7B": 0.14231519234178336,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.04754476156709803
@@ -887,6 +1004,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 36.36015325670498,
             "Qwen2-Audio-7B-Instruct": 41.60919540229885,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 47.356321839080465,
+            "phi_4_multimodal_instruct": 43.524904214559385,
             "WavLLM_fairseq": 41.57088122605364,
             "SALMONN_7B": 30.536398467432953,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 36.81992337164751
@@ -899,6 +1017,7 @@
             "Qwen2-Audio-7B-Instruct": 0.06114048472375004,
             "whisper_large_v3": 0.037649480146197796,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.04900464852205386,
+            "phi_4_multimodal_instruct": 0.028494375643163834,
             "WavLLM_fairseq": 0.06621482559171073,
             "SALMONN_7B": 0.0459884319222171,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.038146268762641496
@@ -912,6 +1031,7 @@
             "whisper_large_v3": 0.7225930420711975,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 0.7824973031283711,
             "gemini-1.5-flash": 0.9690871089536138,
+            "phi_4_multimodal_instruct": 0.7126483279395901,
             "WavLLM_fairseq": 1.2913969795037756,
             "SALMONN_7B": 1.2721817691477886,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 0.6848705501618123
@@ -935,6 +1055,7 @@
             "MERaLiON-AudioLLM-Whisper-SEA-LION": 64.80000000000001,
             "Qwen2-Audio-7B-Instruct": 51.6,
             "cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct": 74.0,
+            "phi_4_multimodal_instruct": 49.0,
             "WavLLM_fairseq": 50.8,
             "SALMONN_7B": 44.6,
             "cascade_whisper_large_v3_llama_3_8b_instruct": 57.800000000000004
@@ -1006,47 +1127,6 @@
             "cascade_whisper_large_v3_llama_3_8b_instruct": 63.68000000000001
         }
     },
-    "imda_part4_30s_asr_test": {
-        "wer": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.0
-        }
-    },
-    "mmau_mini": {
-        "string_match": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 60.5
-        },
-        "llama3_70b_judge": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 64.60000000000001,
-            "phi_4_multimodal_instruct": 59.4
-        }
-    },
-    "mmau_mini_music": {
-        "string_match": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.6077844311377245
-        },
-        "llama3_70b_judge": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.6437125748502994,
-            "phi_4_multimodal_instruct": 0.688622754491018
-        }
-    },
-    "mmau_mini_sound": {
-        "string_match": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.6606606606606606
-        },
-        "llama3_70b_judge": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.7027027027027027,
-            "phi_4_multimodal_instruct": 0.6456456456456456
-        }
-    },
-    "mmau_mini_speech": {
-        "string_match": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.5465465465465466
-        },
-        "llama3_70b_judge": {
-            "MERaLiON-AudioLLM-Whisper-SEA-LION": 0.5915915915915916,
-            "phi_4_multimodal_instruct": 0.44744744744744747
-        }
-    },
     "imda_30s_ar_test": {
         "llama3_70b_judge": {
             "Qwen2-Audio-7B-Instruct": 5.106666666666667,