mkolenic/EuroSpeech-subset
收藏Hugging Face2025-12-11 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/mkolenic/EuroSpeech-subset
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含斯洛伐克配置的音频记录，采样率为16000Hz，以及相关的转录文本（包括自动语音识别（ASR）和人工转录）。数据集还包含国家、语言、视频ID、转录ID、开始和结束时间、持续时间等特征。此外，还提供了字符错误率（CER）和词错误率（WER）等评估指标。数据集分为训练集、验证集和测试集，分别包含640282、6861和5291个样本。
The dataset contains audio recordings for the slovakia configuration with a sampling rate of 16000Hz, along with related transcripts (including ASR and human transcripts). The dataset also includes features such as country, language, video ID, transcript ID, start and end times, duration, etc. Additionally, evaluation metrics like Character Error Rate (CER) and Word Error Rate (WER) are provided. The dataset is divided into train, validation, and test sets with 640282, 6861, and 5291 examples respectively.
提供机构：
mkolenic
