BrunoHays/mixed_multilingual_commonvoice_all_languages_100k

该数据集是从Mozilla Common Voice 13构建的，使用了该仓库中的脚本。数据集的主要用途是教模型忽略非法语的语言。数据集包含音频和句子文本，音频采样率为16000Hz，句子文本为字符串类型，还包括语言区域信息。数据集分为训练集和测试集，训练集包含100,000个样本，测试集包含1,000个样本。

This dataset is built from Mozilla Common Voice 13 using the script committed in this repository. The primary use of the dataset is to teach a model to ignore languages that are not French. The dataset includes audio and sentence text, with audio sampled at 16000Hz and sentence text as string type, along with locale information. The dataset is divided into a training set with 100,000 examples and a test set with 1,000 examples.