Fish-Hebrew

This dataset is a derivative of the gold-standard [Hebrew Speech](https://huggingface.co/datasets/upai-inc/saspeech) dataset. It contains the exact pre-processed files used to train the **Hebrew-Fish** model available in this repository. Additionally, it includes the protobuf file used by Fish during fine-tuning.

本数据集为金标准**希伯来语音（Hebrew Speech）**数据集的衍生数据集，其原始数据集链接为：https://huggingface.co/datasets/upai-inc/saspeech。本数据集包含用于训练本仓库中提供的**Hebrew-Fish**模型的全部精确预处理文件。此外，本数据集还包含了Fish在微调阶段使用的协议缓冲区（protobuf）文件。