The People’s Speech

People's Speech 是一个可免费下载的 30,000 小时且不断增长的受监督会话英语语音识别数据集，根据 CC-BY-SA（带有 CC-BY 子集）许可用于学术和商业用途。这些数据是通过在互联网上搜索具有现有转录的适当许可的音频数据来收集的。

People's Speech is a freely downloadable, 30,000-hour and growing supervised conversational English speech recognition dataset, licensed under CC-BY-SA (with a CC-BY subset) for both academic and commercial use. This dataset is collected by searching the Internet for appropriately licensed audio data with existing transcriptions.

搜集汇总

背景与挑战

背景概述

The People's Speech 是一个可免费下载的大规模英语语音识别数据集，包含超过30,000小时的监督式会话音频，适用于学术和商业用途，基于CC-BY-SA许可。该数据集通过收集互联网上带有转录的许可音频构建，并持续增长，由多个机构于2021年发布。