[
{
"id": "crosstalk_audio_0",
"input_path": "/input/crosstalk/audio_0.mp3",
"text": "请说一段经典的相声，题材可以是关于医生和病人的趣事。",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_0/audio_0.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/00.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_0.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_0.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_0.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_0.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_1",
"input_path": "/input/crosstalk/audio_1.mp3",
"text": "你来扮演捧哏，我说：“这家店的老板怎么这么小气？”，你怎么接话？",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_1/audio_1.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/01.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_1.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_1.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_1.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_1.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_2",
"input_path": "/input/crosstalk/audio_2.mp3",
"text": "现在你来扮演逗哏角色，想象我是捧哏，我说：“今天的天气真不错啊！”，你该怎么接下去？",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_2/audio_2.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/02.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_2.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_2.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_2.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_2.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_3",
"input_path": "/input/crosstalk/audio_3.mp3",
"text": "讲一段融合现代元素（比如互联网语言）的相声，内容关于在网上购物的经历。",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_3/audio_3.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/03.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_3.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_3.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_3.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_3.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_4",
"input_path": "/input/crosstalk/audio_4.mp3",
"text": "我会说出一个情景，你作为相声演员要即兴发挥，继续扩展故事。情景是：在火车站不小心丢了票。",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_4/audio_4.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/04.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_4.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_4.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_4.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_4.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_5",
"input_path": "/input/crosstalk/audio_5.mp3",
"text": "试着讲一段涉及中西文化差异的相声，比如中国的春节与西方的圣诞节的对比。",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_5/audio_5.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/05.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_5.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_5.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_5.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_5.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_6",
"input_path": "/input/crosstalk/audio_6.mp3",
"text": "模仿郭德纲的风格，说一段关于“互联网生活”的相声。",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_6/audio_6.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/06.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_6.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_6.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_6.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_6.wav",
"level": "L2"
},
{
"id": "crosstalk_audio_7",
"input_path": "/input/crosstalk/audio_7.mp3",
"text": "通过相声风格来解释一下：“为什么学习编程这么难？”",
"task": "Crosstalk ability",
"task_description": "Can the model perform a skit, playing both roles in a comedic dialogue?",
"output_path_4o": "/output/ChatGPT-4o/crosstalk/audio_7/audio_7.wav",
"output_path_miniomni": "/output/Mini-Omni/crosstalk/07.wav",
"output_path_speechgpt": "/output/SpeechGPT/crosstalk/answer_7.wav",
"output_path_funaudio": "/output/FunAudioLLM/crosstalk/audio_7.wav",
"language": "Chinese",
"category": "Entertainment",
"output_path_4o_cascade": "/output/cascade/crosstalk/audio_7.wav",
"output_path_4o_llama_omni": "/output/LLaMA_omni/crosstalk/audio_7.wav",
"level": "L2"
}
]