Upload README.md with huggingface_hub

README.md

@@ -5,11 +5,11 @@
 > This dataset only contains test data, which is integrated into UltraEval-Audio(https://github.com/OpenBMB/UltraEval-Audio) framework.
 
 ```shell
-python audio_evals/main.py --dataset tedlium.yaml --model gpt4o_audio
+python audio_evals/main.py --dataset tedlium-release1 --model gpt4o_audio
 
-python audio_evals/main.py --dataset tedlium.yaml --model gpt4o_audio
+python audio_evals/main.py --dataset tedlium-release2 --model gpt4o_audio
 
-python audio_evals/main.py --dataset tedlium.yaml --model gpt4o_audio
+python audio_evals/main.py --dataset tedlium-release3 --model gpt4o_audio
 ```