数据堂—1,441小时意大利语手机采集语音数据

1,441小时意大利语手机采集语音数据由3,109名意大利本土发音人参与录制，意大利口音正宗。录音内容涵盖通用类、交互类、车载命令类、家居命令等多类别，内容丰富。1,441小时意大利语手机采集语音数据的录音文本由语言专家参与设计，人工校对，准确率高。匹配主流安卓、苹果系统手机

This dataset consists of 1,441 hours of Italian speech data collected via mobile phones, recorded by 3,109 native Italian speakers with authentic Italian accents. The recorded content spans multiple categories including general scenarios, interactive scenarios, in-car commands, smart home commands and more, featuring rich and diverse content. The transcriptions of this speech dataset were designed by language experts and manually proofread, ensuring high accuracy. It is compatible with mainstream Android and Apple mobile operating systems.

搜集汇总

背景与挑战

背景概述

该数据集包含1,441小时通过手机采集的意大利语语音数据，由3,109名母语者录制，覆盖对话、命令和新闻等多种内容，专用于意大利语语音识别模型的测试。数据采用16kHz、16位WAV格式，并确保高准确度。