AdoCleanCode/dac_inference_SE2_v3
收藏Hugging Face2025-10-16 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/AdoCleanCode/dac_inference_SE2_v3
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含文本和音频的样本，每个样本有唯一的ID，并分为噪声类型。文本部分包含原始的脏序列、清理后的序列和生成的序列，而音频部分则包含对应的脏音频、干净音频和生成音频。数据集的目的是用于训练模型处理噪声数据，并生成清理后的数据。训练集包含50个示例，整个数据集的大小为52573373字节。
The dataset consists of samples with text and audio, each with a unique ID and categorized by noise type. The text part includes the original dirty sequence, the cleaned sequence, and the generated sequence, while the audio part includes the corresponding dirty audio, clean audio, and generated audio. The dataset is intended for training models to process noisy data and generate cleaned data. The training set contains 50 examples, and the entire dataset is 52573373 bytes in size.
提供机构：
AdoCleanCode
