AudioLLMs/wavcaps_qa_test
收藏Hugging Face2025-03-13 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/AudioLLMs/wavcaps_qa_test
下载链接
链接失效反馈官方服务:
资源简介:
Wavcaps是一个由ChatGPT协助的弱标记音频字幕数据集,用于音频-语言多模态研究。数据集包含音频片段(context)、指令(instruction)和回答(answer)三种类型的数据。目前公开的测试集(test split)包含304个音频示例,数据集总大小为100123494字节。
Wavcaps is a ChatGPT-assisted weakly-labelled audio captioning dataset for audio-language multimodal research. The dataset includes three types of data: audio clips (context), instructions (instruction), and answers (answer). The publicly available test set (test split) contains 304 audio examples, with the total dataset size being 100123494 bytes.
提供机构:
AudioLLMs



