SEACrowd
收藏资源简介:
SEACrowd是一个针对东南亚语言的多语言多模态数据集,由AI新加坡等机构合作创建。该数据集包含近1000种东南亚语言的文本、图像和音频数据,总计498个数据集。创建过程中,通过标准化数据加载器和数据表单,确保数据的可访问性和一致性。SEACrowd旨在解决东南亚语言在AI模型中代表性不足的问题,支持包括自然语言处理、视觉语言和语音识别在内的多种应用,以促进该地区AI技术的发展和语言多样性的保护。
SEACrowd is a multilingual and multimodal dataset focused on Southeast Asian languages, co-created by institutions such as AI Singapore and other collaborating partners. It encompasses text, image and audio data spanning nearly 1,000 Southeast Asian languages, totaling 498 datasets. During its development, standardized data loaders and data sheets were adopted to ensure data accessibility and consistency. SEACrowd aims to address the underrepresentation of Southeast Asian languages in AI models, and supports a range of applications including natural language processing, vision-language tasks and speech recognition, so as to promote the development of AI technology in the region and the preservation of linguistic diversity.

- 1SEACrowd: A Multilingual Multimodal Data Hub and Benchmark Suite for Southeast Asian LanguagesAI Singapore · 2024年



