Wuhan University, in collaboration with China Mobile's Jiutian AI team and Duke Kunshan University, has released the open-source audio-video speaker recognition dataset VoxBlink2, which is based on YouTube data and contains over 110,000 hours of audio-video recordings. The dataset includes 9,904,382 high-quality audio clips and their corresponding video segments, sourced from 111,284 users on YouTube, making it the largest publicly available audio-video speaker recognition dataset to date. The release of this dataset aims to enrich open-source speech corpora and support the training of voiceprint large models.

["Fen Miao Frame is a one-stop audio-video production collaboration platform that provides online review and annotation features.", "The platform aims to improve the efficiency of team revisions and output, while reducing labor costs.", "Features include online review, file sharing, process management, asset management, and more.", "Applicable to various fields such as advertising, media, education, gaming, and film.", "Fen Miao Frame has served over 1 million creators and more than 3,000 enterprises."]