HKUSTAudio/Llasa_opensource_speech_data_160k_hours_tokenized
收藏Hugging Face2025-02-13 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/HKUSTAudio/Llasa_opensource_speech_data_160k_hours_tokenized
下载链接
链接失效反馈官方服务:
资源简介:
该数据集脚本用于合并存储在memmap格式中的标记语音数据集。输入数据集可以组合成更大的训练数据集。该脚本会检查数据集中序列长度的 consistency,合并数据,并以memmap格式保存新的合并后的数据集。
This dataset script is used for merging tokenized speech datasets stored in memmap format. Input datasets can be combined to form larger training datasets. The script checks for sequence length consistency across datasets, merges the data, and saves the new merged dataset in memmap format.
提供机构:
HKUSTAudio



