CSRC (Children Speech Recognition Challenge)

儿童语音识别对于开发儿童机器交互和计算机辅助语言学习等智能技术至关重要。但是儿童讲话的声学和语言模式与成人讲话的声学和语言模式明显不同。此外，如今儿童语音数据集仍然有限，这可能会阻碍儿童语音识别研究的发展。对此，我们推出了儿童语音识别挑战赛，作为 IEEE SLT 2021 研讨会的旗舰卫星赛事，将为注册团队发布约 400 小时的数据，并设置两个挑战赛道，旨在推动儿童语音识别研究。希望挑战赛也能为领域适应、迁移学习等相关技术提供良好的测试平台。

Child speech recognition is critical for developing intelligent technologies such as child-machine interaction and computer-assisted language learning. However, the acoustic and linguistic patterns of children’s speech differ significantly from those of adult speech. Furthermore, the current availability of child speech datasets remains limited, which may hinder the advancement of child speech recognition research. To address this issue, we launch the Child Speech Recognition Challenge, a flagship satellite event of the IEEE SLT 2021 workshop. The challenge will release approximately 400 hours of data for registered teams and feature two challenge tracks, aiming to promote research in child speech recognition. It is also hoped that this challenge will serve as an excellent testbed for related technologies such as domain adaptation and transfer learning.

搜集汇总

背景与挑战

背景概述

CSRC数据集是IEEE SLT 2021研讨会的儿童语音识别挑战赛数据，提供约400小时的儿童语音数据，旨在推动儿童语音识别研究并作为领域适应等相关技术的测试平台。