linagora/linto-dataset-audio-ar-tn-augmented

LinTO阿拉伯突尼斯方言音频增强数据集是一个适用于阿拉伯突尼斯方言自动语音识别任务的数据集。它基于原始的LinTO阿拉伯突尼斯方言音频数据集，并对其进行了噪声降低和语音转换等增强处理。数据集包含来自不同来源和类型的音频，如足球新闻、纪录片、播客、作者朗读、学习资源和混合主题等，支持阿拉伯突尼斯方言、法语和英语的代码切换。

The LinTO DataSet Audio for Arabic Tunisian Augmented is a dataset tailored for Arabic Tunisian automatic speech recognition (ASR) tasks. It builds on the original LinTO DataSet Audio for Arabic Tunisian, applying noise reduction and voice conversion techniques to enhance the dataset. The dataset includes a variety of audio sources and types such as football news, documentaries, podcasts, authors readings, learning resources, and mixed topics, supporting code-switching between Tunisian Arabic, French, and English.