results/wer/asr.csv

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,imda_part3_asr_test,imda_part4_asr_test,imda_part5_asr_test,imda_part6_asr_test
salmonn_7b,0.555834307918663,0.4180490665254317,0.3375363911542985,0.34335568653146914,0.14226764839341272,0.2687240831119165,0.3638199574803293,0.08558958163301515,0.18386810849140223,0.1581255622670608,0.6915082052617869,0.8632952723448728,1.0998724881343125,0.7462024043822395,0.5240368577507283
wavllm_fairseq,0.02103218017882069,0.04798834811886432,0.14533325621300636,0.3792176325635977,0.15491778414546403,0.6447482518259942,0.6671766188447099,0.06621482559171073,0.4536784258110264,0.07762498393522684,0.4511070591299818,0.8449539438237986,,0.652558845214085,0.5275247040334478
Qwen2-Audio-7B-Instruct,0.03201041234390915,0.06073732195887794,0.11438872500819404,0.22323599266520214,0.11891779417331638,0.9865371146513101,0.9884312836905533,0.06387282211419042,0.953518879631271,,,,,,
whisper_large_v3_with_llama_3_8b_instruct,0.01831591655034519,0.03714982881570734,0.09876543209876543,0.14540692118393275,0.09515429104337297,0.11773910240019567,0.15611126487402763,0.038146268762641496,0.04754476156709803,0.0464432592211798,0.2904141703568638,0.33025751212260873,0.5142358733382134,0.4662862470067086,0.28367569330674514
qwen_audio_chat,0.02246576376051609,0.04163277659031154,0.11645169375518152,0.30723956433545313,0.13316235845059654,0.38460383986837277,0.511809498898483,0.03995599872254356,0.45289842226555577,0.08434005911836524,0.5057046105756707,1.0992041596614108,1.3267786724597983,0.7411298037375774,0.46036235053411123