Speech Accent Archive

建立语音口音档案库是为了统一展示来自各种语言背景的大量语音口音。讲英语的母语和非母语人士都阅读相同的英语段落，并进行了认真记录。存档被构造为教学工具和研究工具。它旨在供语言学家以及其他只希望听和比较不同英语使用者的口音的人使用。
通过此数据集，您可以比较说话者的人口统计学和语言背景，以确定哪些变量是每种口音的关键预测指标。语音重音档案库表明，重音是系统性的，而不仅仅是错误的语音。

The Speech Accent Archive was established to uniformly showcase a large collection of speech accents from diverse linguistic backgrounds. Native and non-native English speakers read the same English passages and were carefully recorded. The archive is structured as both a teaching and research tool, intended for use by linguists as well as others who simply wish to listen to and compare the accents of different English speakers.
With this dataset, users can compare the demographic and linguistic backgrounds of speakers to identify which variables serve as key predictive indicators for each accent. The Speech Accent Archive demonstrates that accents are systematic, not merely erroneous speech.

搜集汇总

背景与挑战

背景概述

Speech Accent Archive是一个用于展示多种语言背景语音口音的数据集，包含来自177个国家、214种母语的说话者朗读相同英语段落的2140个录音样本。它作为教学和研究工具，旨在帮助用户分析和比较不同英语使用者的口音，数据文件包括文本、人口统计信息和音频，采用CC BY-NC-SA 4.0许可协议。