MALIBA-AI/bambara-speech-recognition-leaderboard

这是一个包含音频和文本信息的数据集，分为eval和preview两个部分。eval部分包含568个音频样本，preview部分包含113个音频和文本样本。每个样本都有唯一的id、音频文件、文本内容（仅限于preview部分）、类别和持续时间信息。音频采样率为16000Hz。数据集遵循Apache-2.0许可。

This dataset includes audio and text information, divided into eval and preview sections. The eval section contains 568 audio samples, while the preview section contains 113 audio and text samples. Each sample has a unique id, audio file, text content (limited to the preview section), category, and duration information. The audio sampling rate is 16000Hz. The dataset follows the Apache-2.0 license.