kidsNARRATE

kidsNARRATE是一个专为研究中国-英语双语学龄前儿童第二语言叙事能力而设计的非母语语音语料库。该数据集包含6小时的儿童叙事测试音频记录，以及人工转录、评分和词级语法及发音错误标注。数据集还提供了视频辅助转录，以提高准确性。此外，数据集处理了音频信号，以满足机器学习应用的需求。kidsNARRATE不仅适用于儿童第二语言习得研究，也支持自动语音识别技术的发展，旨在解决双语儿童叙事能力评估的挑战。

kidsNARRATE is a non-native speech corpus specifically designed for researching the second language narrative abilities of Chinese-English bilingual preschool children. This dataset contains 6 hours of audio recordings from children's narrative tests, along with manual transcriptions, scoring, and word-level grammatical and pronunciation error annotations. It also provides video-assisted transcriptions to enhance transcription accuracy. Additionally, the audio signals within the dataset have been preprocessed to meet the requirements of machine learning applications. kidsNARRATE is applicable not only to research on children’s second language acquisition but also supports the development of automatic speech recognition technologies, aiming to address the challenges in evaluating the narrative abilities of bilingual children.