[ { "id": "story_audio_0", "input_path": "/input/story/audio_0.mp3", "text": "Tell a sad story, such as the journey of a little kitten being abandoned and later adopted by a kind person.", "task": "Storytelling", "task_description": "Can the model narrate a story with emotional depth?", "output_path_4o": "/output/ChatGPT-4o/story/audio_0/audio_0.wav", "output_path_miniomni": "/output/Mini-Omni/story/00.wav", "output_path_speechgpt": "/output/SpeechGPT/story/answer_0.wav", "output_path_funaudio": "/output/FunAudioLLM/story/audio_0.wav", "text_cn": "讲一个悲伤的故事,例如一只小猫被抛弃并后来被一个善良的人采用的旅程。", "language": "English", "category": "Entertainment", "output_path_4o_cascade": "/output/cascade/story/audio_0.wav", "output_path_4o_llama_omni": "/output/LLaMA_omni/story/audio_0.wav", "level": "L2" }, { "id": "story_audio_1", "input_path": "/input/story/audio_1.mp3", "text": "Begin a mysterious story set in an ancient, abandoned castle.", "task": "Storytelling", "task_description": "Can the model narrate a story with emotional depth?", "output_path_4o": "/output/ChatGPT-4o/story/audio_1/audio_1.wav", "output_path_miniomni": "/output/Mini-Omni/story/01.wav", "output_path_speechgpt": "/output/SpeechGPT/story/answer_1.wav", "output_path_funaudio": "/output/FunAudioLLM/story/audio_1.wav", "text_cn": "在一个古老的废弃城堡中开始一个神秘的故事。", "language": "English", "category": "Entertainment", "output_path_4o_cascade": "/output/cascade/story/audio_1.wav", "output_path_4o_llama_omni": "/output/LLaMA_omni/story/audio_1.wav", "level": "L2" }, { "id": "story_audio_2", "input_path": "/input/story/audio_2.mp3", "text": "Describe a story where a little boy discovers a lost puppy in the snow, detailing their first meeting.", "task": "Storytelling", "task_description": "Can the model narrate a story with emotional depth?", "output_path_4o": "/output/ChatGPT-4o/story/audio_2/audio_2.wav", "output_path_miniomni": "/output/Mini-Omni/story/02.wav", "output_path_speechgpt": "/output/SpeechGPT/story/answer_2.wav", "output_path_funaudio": "/output/FunAudioLLM/story/audio_2.wav", "text_cn": "描述一个故事,一个小男孩在雪中发现一只迷失的小狗,详细介绍了他们的第一次见面。", "language": "English", "category": "Entertainment", "output_path_4o_cascade": "/output/cascade/story/audio_2.wav", "output_path_4o_llama_omni": "/output/LLaMA_omni/story/audio_2.wav", "level": "L2" }, { "id": "story_story0_CopyMachine_1", "input_path": "/input/noise/story0_CopyMachine_1.wav", "text": "[Add CopyMachine noise]Tell a sad story, such as the journey of a little kitten being abandoned and later adopted by a kind person.", "noise":"Add CopyMachine noise", "task": "Storytelling", "task_description": "Can the model narrate a story with emotional depth?", "output_path_4o": "/output/ChatGPT-4o/noise/story0_CopyMachine_1/story0_CopyMachine_1.wav", "output_path_miniomni": "/output/Mini-Omni/noise/10.wav", "output_path_speechgpt": "/output/SpeechGPT/noise/story0_CopyMachine_1.wav", "output_path_funaudio": "/output/FunAudioLLM/noise/audio_10.wav", "text_cn": "讲一个悲伤的故事,例如一只小猫被抛弃并后来被一个善良的人采用的旅程。", "language": "English", "category": "Entertainment", "output_path_4o_cascade": "/output/cascade/noise/story0_CopyMachine_1.wav", "output_path_4o_llama_omni": "/output/LLaMA_omni/noise/story0_CopyMachine_1.wav", "level": "L2" }, { "id": "story_story1_CopyMachine_1", "input_path": "/input/noise/story1_CopyMachine_1.wav", "text": "[Add CopyMachine noise]Begin a mysterious story set in an ancient, abandoned castle.", "noise":"Add CopyMachine noise", "task": "Storytelling", "task_description": "Can the model narrate a story with emotional depth?", "output_path_4o": "/output/ChatGPT-4o/noise/story1_CopyMachine_1/story1_CopyMachine_1.wav", "output_path_miniomni": "/output/Mini-Omni/noise/11.wav", "output_path_speechgpt": "/output/SpeechGPT/noise/story1_CopyMachine_1.wav", "output_path_funaudio": "/output/FunAudioLLM/noise/audio_11.wav", "text_cn": "在一个古老的废弃城堡中开始一个神秘的故事。", "language": "English", "category": "Entertainment", "output_path_4o_cascade": "/output/cascade/noise/story1_CopyMachine_1.wav", "output_path_4o_llama_omni": "/output/LLaMA_omni/noise/story1_CopyMachine_1.wav", "level": "L2" } ]