chiyuanhsiao/audio_merge-linear_trivia_qa-audio

该数据集包含了问题、问题ID、问题来源、实体页面信息（包括文档来源、文件名、标题、维基百科上下文）、搜索结果（包括描述、文件名、排名、标题、URL、搜索上下文）、答案（包括别名、标准化别名、匹配的维基实体名称、标准化的匹配维基实体名称、标准化值、类型、值）、问题单位、响应交错的文本、响应文本、响应令牌数量和响应语音等字段。数据集包含一个验证集，大小为943211424字节，共有1000个示例。

The dataset includes fields such as questions, question IDs, question sources, entity page information (including document sources, filenames, titles, Wikipedia contexts), search results (including descriptions, filenames, ranks, titles, URLs, search contexts), answers (including aliases, normalized aliases, matched Wikipedia entity names, normalized matched Wikipedia entity names, normalized values, types, and values), question units, interleaved response text, response text, number of response tokens, and response audio. The dataset contains a validation set, which is 943211424 bytes in size and includes 1000 examples.