disco-eth/AIME

AIME（AI音乐评估）数据集包含了由12种音乐生成模型生成的6000个音频轨道，以及来自MTG-Jamendo数据集的500个轨道。用于生成音乐的提示是MTG-Jamendo数据集中具有代表性的多样化标签的组合。该数据集包含两个子集：AIME音频数据集和AIME调查数据集。数据集中的字段包括：id（用于将调查数据映射到音频文件）、model（生成音频样本的模型名称）、description（包含三个字符串，列出用于提示音乐生成模型的标签）和audio（包含音频样本）。

The AIME (AI Music Evaluation) dataset consists of 6000 audio tracks generated by 12 different music generation models, as well as 500 tracks from the MTG-Jamendo dataset. The prompts used to generate the music are combinations of representative and diverse tags from the MTG-Jamendo dataset. The dataset includes two subsets: the AIME audio dataset and the AIME survey dataset. The fields in the dataset include: id (used to map survey data to audio files), model (the name of the model used to generate the audio sample), description (a list of three strings containing the tags used to prompt the music generation model), and audio (contains the audio samples).