lmms-lab/alpaca_audio
收藏Hugging Face2024-10-17 更新2025-04-08 收录
下载链接:
https://hf-mirror.com/datasets/lmms-lab/alpaca_audio
下载链接
链接失效反馈官方服务:
资源简介:
这是一个包含音频数据和对应文本指令及答案的数据集。音频数据的采样率为192000,每个样本包括一个音频上下文字段(context)、一个文本指令(instruction)、一个语音指令(speech_instruction)和一个答案(answer)。数据集分为测试集,共有100个样本。
This dataset consists of audio data and corresponding text instructions and answers. The audio data has a sampling rate of 192000, and each sample includes an audio context field, a text instruction, a speech instruction, and an answer. The dataset is split into a test set with a total of 100 samples.
提供机构:
lmms-lab



