magicdata-ramc

为了进一步丰富开源语音语料库，促进语音语言处理技术的发展，Magic Data联合中科院声学研究所、上海交通大学和西北工业大学，在Magichub开源社区正式开源180小时中文对话式语音数据集MagicData-RAMC。MagicData-RAMC是一批高质量且标注丰富的训练数据，可以很好地支持开发者完成语音识别和说话人日志相关的研究。数据集下载地址https://magichub.com/datasets/magicdata-ramc/To further enrich open-source speech corpora and advance the development of speech and language processing technologies, Magic Data, in collaboration with the Institute of Acoustics of the Chinese Academy of Sciences, Shanghai Jiao Tong University, and Northwestern Polytechnical University, officially open-sourced the 180-hour Chinese conversational speech dataset MagicData-RAMC on the Magichub open-source community. MagicData-RAMC is a high-quality, richly annotated training dataset that can effectively support researchers and developers in conducting studies related to speech recognition and speaker diarization. The download link of the dataset is https://magichub.com/datasets/magicdata-ramc/

搜集汇总

背景与挑战

背景概述

MagicData-RAMC是一个180小时的中文对话式语音数据集，由Magic Data联合多所研究机构在Magichub开源社区发布。该数据集以高质量和丰富标注为特点，专为支持语音识别和说话人日志研究而设计。