amuvarma/conversation-elias-3-0-t120-convo-both-full-raw-fac

该数据集包含多个用户和助手的文本和音频数据。具体特征包括用户和助手的文本（user_1_text到user_6_text和assistant_1_text到assistant_6_text），以及对应的音频URL和音频文件（assistant_1_audiourl到assistant_6_audiourl和assistant_1_audio到assistant_6_audio）。此外，还包含用户文本对应的音频数据（user_1_text_audio到user_6_text_audio），以及一些音频编码数据（user_6_text_audio_facodec_0到user_6_text_audio_facodec_5）。数据集分为一个训练集（train），包含119个样本，总大小为226035402字节。

This dataset contains text and audio data from multiple users and assistants. The features include text from users and assistants (user_1_text to user_6_text and assistant_1_text to assistant_6_text), as well as corresponding audio URLs and audio files (assistant_1_audiourl to assistant_6_audiourl and assistant_1_audio to assistant_6_audio). Additionally, it includes audio data corresponding to user text (user_1_text_audio to user_6_text_audio), and some audio encoding data (user_6_text_audio_facodec_0 to user_6_text_audio_facodec_5). The dataset is divided into a training set (train) containing 119 samples, with a total size of 226035402 bytes.