Classroom Explanation Chinese Voice Content Classification Dataset

In the education and training industry, with the booming development of online education and digital classrooms, a vast amount of voice explanation content needs to be efficiently managed and analyzed, which is a major challenge currently faced. Existing audio content classification methods typically rely on manual labeling and analysis, which are inefficient and prone to bias. This dataset aims to provide high-quality voice data to solve the technical challenges of automatic classification and analysis of voice content, meeting the business needs of precise educational content distribution and intelligent learning assistance. Data collection uses high-sensitivity microphones and professional recording equipment in real classroom environments. To ensure high quality of voice data, we used multiple rounds of annotation and consistency checks, with a labeling team consisting of linguistics experts and education specialists, scaling up to 20 people. Data preprocessing includes noise elimination, voice segmentation, and feature extraction, stored in WAV format, combined with JSON files to record annotation information and metadata.