Fisher English Training Speech Part 1 Transcripts

<h3>Introduction</h3><br>
<p>Fisher English Training Speech Part 1 Transcripts represents the first half of a collection of conversational telephone speech (CTS) that was created at LDC in 2003. It contains time-aligned transcript data for 5,850 complete conversations, each lasting up to 10 minutes. In addition to the transcriptions, which are found under the trans directory, there is a complete set of tables describing the speakers, the properties of the telephone calls, and the set of topics that were used to initiate the conversations. The corresponding speech files are contained in <a href="http://catalog.ldc.upenn.edu/LDC2004S13" rel="nofollow">Fisher English Training Speech Part 1 Speech (LDC2004S13)</a>.</p><br>
<p>The Fisher telephone conversation collection protocol was created at LDC to address a critical need of developers trying to build robust automatic speech recognition (ASR) systems. Previous collection protocols, such as CALLFRIEND and Switchboard-II and the resulting corpora, have been adapted for ASR research but were in fact developed for language and speaker identification respectively. Although the CALLHOME protocol and corpora were developed to support ASR technology, they feature small numbers of speakers making telephone calls of relatively long duration with narrow vocabulary across the collection. CALLHOME conversations are challengingly natural and intimate. Under the Fisher protocol, a large number of participants each calls an other participant, whom they typically do not know, for a short short period of time to discuss the assigned topics.&nbsp;This maximizes inter-speaker variation and vocabulary breath while also increasing formality.</p><br>
<p>Previous protocols such as CALLHOME, CALLFRIEND and Switchboard relied upon participant activity to drive the collection. Fisher is unique in being platform driven rather than participant driven. Participants who wish to initiate a call may do so, however, the collection platform initiates the majority of calls. Participants need only answer their phones at the times they specified when registering for the study.</p><br>
<p>To encourage a broad range of vocabulary, Fisher participants are asked to speak about an assigned topic chosen from a randomly generated list that changes every 24 hours. All participants that day will be assigned subjects from that list.&nbsp;Some topics are inherited or refined from previous Switchboard studies while others were developed specifically for the Fisher protocol.</p><br>
<h3>Data</h3><br>
<p>Overall, about 12% of the conversations were transcribed at LDC, and the rest were transcribed by BBN and WordWave using a significantly different approach to the task. A central goal in both sets was to maximize the speed and economy of the transcription process. This in turn involved certain aspects of mark-up detail and quality control that may have been common in previous, smaller corpora.</p><br>
<p>The LDC transcripts were based on automatic segmentation of the audio data, to identify the utterance end-points on both channels of each conversation. Given these time stamps, manual transcription was simply a matter of typing in the words for each segment and doing a rudimentary spell-check. No attempt was made to modify the segmentation boundaries manually, or to locate utterances that the segmenter might have missed. Portions of speech where the transcriber could not be sure exactly what was said were marked with double parentheses -- (( ... )) -- and the transcriber could hazard a guess as to what was said, or leave the region between parentheses blank. The LDC transcription process yields one plain-text transcript file per conversation, in which the first two lines show the call-ID and the fact that the transcript was developed at LDC. The remainder of the file contains one utterance per line (with blank lines separating the utterances), with the start-time, end-time, speaker/channel-ID and utterance text.</p><br>
<p>Data collection and transcription were sponsored by DARPA and the U.S. Department of Defense, as part of the EARS project for research and development in automatic speech recognition.</p><br>
<h3>Samples</h3><br>
<p>Please examine this <a href="desc/addenda/LDC2004T19.txt" rel="nofollow">sample</a> to see an example of the data in this corpus.</p><br>
<h3>Updates</h3><br>
<p>As of 6/14/2017, 'fe_03_p1_calldata.tbl' was updated to correct mislabeled topics for some calls. All&nbsp;downloads made after this date will have the corrected file.</p></br>
© 2003-2004 Trustees of the University of Pennsylvania

<h3>引言</h3><br><p>Fisher英语训练语音第一部分转录本是2003年由语言数据联盟（Linguistic Data Consortium，LDC）制作的会话电话语音（Conversational Telephone Speech，CTS）数据集的前半部分。该数据集包含5850段完整对话的时间对齐转录数据，每段对话时长不超过10分钟。除存放在trans目录下的转录文本外，数据集还包含一套完整的表格，用于描述对话参与者、电话通话属性以及用于发起对话的话题集合。对应的语音文件可参见<a href="http://catalog.ldc.upenn.edu/LDC2004S13" rel="nofollow">Fisher英语训练语音第一部分语音数据集（LDC2004S13）</a>。</p><br><p>Fisher电话对话采集协议由LDC制定，旨在满足研发鲁棒性自动语音识别（Automatic Speech Recognition，ASR）系统的开发者们的迫切需求。此前的采集协议如CALLFRIEND、Switchboard-II及其对应的语料库，虽已适配ASR研究，但实际上分别是为语言识别和说话人识别开发的。尽管CALLHOME协议及其语料库是为支持ASR技术研发而打造的，但它们存在以下局限：参与对话的说话者数量较少，通话时长相对较长，且整个数据集的词汇范围较窄。CALLHOME的对话风格极具自然性与私密性，难以建模。在Fisher协议下，大量参与者会拨打彼此的电话（通常双方互不相识），进行短时长的指定话题讨论。该设计最大化了说话人间的差异与词汇广度，同时提升了对话的正式程度。</p><br><p>此前的采集协议如CALLHOME、CALLFRIEND和Switchboard均依赖参与者主动发起通话来完成数据采集。Fisher数据集的独特之处在于其采用平台驱动模式，而非参与者驱动模式。尽管希望发起通话的参与者可主动拨打，但绝大多数通话均由采集平台发起，参与者仅需在注册研究时指定的时段接听电话即可。</p><br><p>为拓展词汇覆盖范围，Fisher数据集的参与者需围绕随机生成的话题列表中的指定话题展开讨论，该话题列表每24小时更新一次。当日所有参与者的话题均取自当日的话题列表。部分话题继承自此前的Switchboard研究并经过优化，其余话题则专为Fisher协议开发。</p><br><h3>数据</h3><br><p>总体而言，约12%的对话由LDC完成转录，其余对话由BBN和WordWave采用截然不同的转录方案完成。两类转录工作的核心目标均为最大化转录流程的速度与经济性，因此在标注细节与质量控制方面采取了部分在过往小型语料库中较为常见的简化措施。</p><br><p>LDC的转录工作基于音频数据的自动分割，用于识别每段对话双声道的语句端点。在获得这些时间戳后，手动转录仅需为每个片段输入文本并进行基础的拼写检查即可。工作人员不会手动修改分割边界，也不会补充分段器遗漏的语句。若转录人员无法确定某段语音的准确内容，将使用双层括号((...))进行标记，此时转录人员可尝试猜测内容，也可将括号内区域留空。LDC的转录流程为每段对话生成一个纯文本转录文件，文件前两行分别标注通话ID以及“本转录本由LDC开发”的说明。文件剩余部分每行对应一条语句（语句间以空行分隔），每行包含起始时间、结束时间、说话者/声道ID以及语句文本。</p><br><p>数据采集与转录工作由美国国防高级研究计划局（Defense Advanced Research Projects Agency，DARPA）与美国国防部资助，作为自动语音识别研发的EARS项目的一部分。</p><br><h3>示例</h3><br><p>您可查看<a href="desc/addenda/LDC2004T19.txt" rel="nofollow">本示例</a>以了解该语料库的数据格式与样例。</p><br><h3>更新记录</h3><br><p>截至2017年6月14日，我们已对'fe_03_p1_calldata.tbl'文件进行更新，修正了部分通话的话题标注错误。该日期之后的所有下载版本均包含修正后的文件。</p><br>© 2003-2004 宾夕法尼亚大学托管委员会

搜集汇总

背景与挑战

背景概述

Fisher English Training Speech Part 1 Transcripts是由LDC开发的英语电话对话转录数据集，包含5,850个对话（984小时）的文本转录和元数据表，用于语音识别研究。该数据集是Fisher集合的第一部分，采用平台驱动收集方式，参与者讨论每日更新的随机话题，以增强词汇多样性和形式性，支持ASR系统开发。