IARPA Babel Cantonese Language Pack IARPA-babel101b-v0.4c

<h3>Introduction</h3><br>
<p>IARPA Babel Cantonese Language Pack IARPA-babel101b-v0.4c was developed by <a href="http://www.appen.com/">Appen</a> for the IARPA (Intelligence Advanced Research Projects Activity) <a href="http://www.iarpa.gov/index.php/research-programs/babel">Babel</a> program. It contains approximately 215 hours of Cantonese conversational and scripted telephone speech collected in 2011 along with corresponding transcripts.</p><br>
<p>The Babel program focuses on underserved languages and seeks to develop speech recognition technology that can be rapidly applied to any human language to support keyword search performance over large amounts of recorded speech.</p><br>
<h3>Data</h3><br>
<p>The Cantonese speech in this release represents that spoken in the Chinese provinces of Guangdong and Guangxi, and within those provinces, among five dialect groups. The gender distribution among speakers is approximately even; speakers' ages range from 16 years to 67 years. Calls were made using different telephones (e.g., mobile, landline) from a variety of environments including the street, a home or office, a public place, and inside a vehicle.</p><br>
<p>All audio data is presented as 8kHz 8-bit a-law encoded audio in sphere format. Transcripts are available in two versions: simplified Chinese characters and a romanization scheme based on the Yale system, both encoded in UTF-8. Further information about transcription methodology is contained in the documentation accompanying this release.</p><br>
<p>Evaluation data is available from <a href="http://www.nist.gov/">NIST</a> in support of <a href="http://www.nist.gov/itl/iad/mig/openkws.cfm">OpenKWS</a>.</p><br>
<h3>Samples</h3><br>
<p>Please view the following samples:</p><br>
<ul><br>
<li><a href="desc/addenda/LDC2016S02.sph">audio sample</a></li><br>
<li><a href="desc/addenda/LDC2016S02.txt">transcription</a></li><br>
<li><a href="desc/addenda/LDC2016S02.rom.txt">romanized transcription</a></li><br>
</ul><br>
<h3>Updates</h3><br>
<p>None at this time.</p></br>
Portions © 2015 U.S. Government
</br>The U.S. Government acquired this data from Appen Pty Ltd, which assigned the copyright to the data to the U.S. Government.

<h3>简介</h3><br>
<p>本数据集为IARPA Babel粤语语言包IARPA-babel101b-v0.4c，由Appen为IARPA（情报高级研究计划局，Intelligence Advanced Research Projects Activity）的Babel项目开发。该数据集包含2011年采集的约215小时粤语会话与脚本式电话语音数据，以及对应的转写文本。</p><br>
<p>Babel项目聚焦服务资源匮乏的语言，旨在研发可快速应用于任意人类语言的语音识别技术，以支持大规模录制语音中的关键词检索任务。</p><br>
<h3>数据说明</h3><br>
<p>本发布的粤语语音数据采集自中国广东省与广西壮族自治区，覆盖上述区域内的五大方言群体。说话人性别分布大致均衡，年龄跨度为16岁至67岁。通话通过不同类型的电话（如移动电话、固定电话）完成，采集环境涵盖街道、家庭或办公室、公共场所及车辆内部等多种场景。</p><br>
<p>所有音频数据均采用sphere格式，以8kHz、8位a-law编码存储。转写文本提供两种版本：简体中文字符与基于耶鲁（Yale）系统的罗马音标注，均以UTF-8编码格式保存。关于转写方法的详细说明，请参阅本发布包附带的文档。</p><br>
<p>评估数据可从NIST（美国国家标准与技术研究院，National Institute of Standards and Technology）获取，用于支持OpenKWS（开放关键词检索）任务。</p><br>
<h3>示例</h3><br>
<p>请查看以下示例：</p><br>
<ul><br>
<li><a href="desc/addenda/LDC2016S02.sph">音频示例</a></li><br>
<li><a href="desc/addenda/LDC2016S02.txt">转写文本</a></li><br>
<li><a href="desc/addenda/LDC2016S02.rom.txt">罗马音标注转写</a></li><br>
</ul><br>
<h3>更新记录</h3><br>
<p>目前暂无更新记录。</p><br>
<p>部分内容 © 2015 美国政府。</p><br>
<p>美国政府从Appen Pty Ltd获取本数据，Appen Pty Ltd已将该数据的版权转让予美国政府。</p>

搜集汇总

背景与挑战

背景概述

IARPA Babel粤语语言包包含215小时粤语电话语音及对应文本，数据来自中国广东和广西地区，涵盖五种方言群体，适用于语音识别研究。音频为8kHz a-law编码，文本提供简中和罗马化版本。