数据堂—1,012小时印度英语手机采集语音数据

1,012小时印度英语手机采集语音数据是由2,100名印度本土发音人参与录制；录音文本由语言专家参与设计，涵盖通用、交互、车载、家居等多类别；文本经过人工校对，准确率高；本套印度英语手机采集语音数据可应用于语音识别、机器翻译、声纹识别

This dataset contains 1,012 hours of Indian English speech data collected via mobile phones, with 2,100 native Indian speakers participating in the recording. The accompanying recording texts were designed by language experts, covering various scenarios such as general daily use, interactive communication, in-vehicle environments, and smart home scenarios. All texts have undergone manual proofreading to ensure high accuracy. This Indian English mobile-collected speech dataset can be applied to tasks including speech recognition, machine translation, and speaker verification.

搜集汇总

背景与挑战

背景概述

该数据集包含1,012小时由2,100名印度母语者通过手机采集的英语语音数据，覆盖通用、交互、车载及家庭命令等多个类别，文本经人工校对确保高准确率。它专为印度英语识别模型测试设计，适用于语音识别、机器翻译和说话人验证等任务。