2008 NIST Speaker Recognition Evaluation Supplemental Set

Introduction 2008 NIST Speaker Recognition Evaluation Supplemental Set, Linguistic Data Consortium (LDC) catalog number LDC2011S11 and ISBN 1-58563-601-0, was developed by LDC and NIST (National Institute of Standards and Technology) and contains additional data distributed after the main 2008 Speaker Recognition Evaluation (SRE). Specifically, the corpus consists of 770 hours of English microphone speech along with transcripts and other materials used as supplemental data in the 2008 NIST Speaker Recognition Evaluation (SRE) and in a follow-up evaluation to SRE08. NIST SRE is part of an ongoing series of evaluations conducted by NIST. These evaluations are an important contribution to the direction of research efforts and the calibration of technical capabilities. They are intended to be of interest to all researchers working on the general problem of text independent speaker recognition. To this end the evaluation is designed to be simple, to focus on core technology issues, to be fully supported, and to be accessible to those wishing to participate. The 2008 evaluation was distinguished from prior evaluations, in particular those in 2005 and 2006, by including not only conversational telephone speech data but also conversational speech data of comparable duration recorded over a microphone channel involving an interview scenario.The follow-up evaluation focused on speaker detection in the context of conversational interview type speech and was designed to measure the performance of SRE08 systems in previously unexposed test segment channel conditions. LDC previously released the main 2008 NIST SRE Evaluation in three parts as 2008 NIST Speaker Recognition Evaluation Training Set Part 1 LDC2011S05, 2008 NIST Speaker Recognition Evaluation Training Set Part 2 LDC2011S07 and 2008 NIST Speaker Recognition Evaluation Test Set LDC2011S08. Additional documentation is available in the 2008 SRE Evaluation Plan and the Plan for Follow-up Evaluation to SRE08. Data The speech data in this release was collected in 2007 by LDC at its Human Subjects Data Collection Laboratories in Philadelphia and by the International Computer Science Institute (ICSI) at the University of California, Berkeley. This collection was part of the Mixer 5 project, which was designed to support the development of robust speaker recognition technology by providing carefully collected and audited speech from a large pool of speakers recorded simultaneously across numerous microphones and in different communicative situations and/or in multiple languages. Mixer participants were native English and bilingual English speakers. The microphone speech in this corpus is in English and consists of approximately 3 minute and 30 minute interview excerpts.. This supplemental data is split into four different parts which provide: new training data distributed to 2008 SRE participants additional data distributed to participants in the 2008 SRE follow-up evaluation interviewer channel files for the 2008 SRE main test (released after the evaluations) supplemental training data (released after the evaluations) English language transcripts in .cfm format were produced using an automatic speech recognition (ASR) system and are included for some, but not all, speech data. Samples For an example of the data contained in this corpus, review this audio sample. Updates None at this time. Portions © 2007, 2011 Trustees of the University of Pennsylvania

2008年NIST说话人识别评测补充集（2008 NIST Speaker Recognition Evaluation Supplemental Set）由语言数据联盟（Linguistic Data Consortium，简称LDC）与美国国家标准与技术研究院（National Institute of Standards and Technology，简称NIST）联合开发，其LDC目录号为LDC2011S11，国际标准书号为ISBN 1-58563-601-0。本数据集包含主版2008年NIST说话人识别评测（Speaker Recognition Evaluation，简称SRE）发布后追加分发的补充数据。

具体而言，该语料库包含770小时英语麦克风语音数据，附带转写文本及其他配套材料，用作2008年NIST说话人识别评测（以下简称SRE08）及其后续评测的补充数据源。NIST SRE是NIST持续开展的系列评测活动之一，此类评测为研究方向指引与技术能力校准提供了重要支撑，旨在吸引所有致力于解决与文本无关的说话人识别通用问题的研究者参与。为此，本次评测设计简洁聚焦核心技术问题，且提供完整支持，便于所有意向参与者获取使用。

与2005、2006年的往届评测相比，2008年的评测颇具特色：不仅包含会话电话语音数据，还收录了同等时长、通过麦克风信道采集的访谈场景会话语音数据。后续评测则聚焦会话访谈类语音场景下的说话人检测任务，旨在评估SRE08系统在此前未涉及的测试段信道条件下的表现。

LDC此前已分三部分发布主版2008年NIST SRE评测数据集：2008 NIST说话人识别评测训练集第1部分（LDC2011S05）、2008 NIST说话人识别评测训练集第2部分（LDC2011S07）以及2008 NIST说话人识别评测测试集（LDC2011S08）。更多官方文档可参阅《2008年SRE评测方案》与《SRE08后续评测方案》。

本次发布的语音数据由LDC于2007年在其费城的人类受试者数据采集实验室，以及加州大学伯克利分校国际计算机科学研究所（International Computer Science Institute，简称ICSI）联合采集。该采集项目隶属于Mixer 5项目，该项目旨在通过从大量受试者群体中精心采集并审核多麦克风信道、多沟通场景乃至多语言的语音数据，为鲁棒性说话人识别技术的研发提供支撑。参与Mixer项目的受试者为英语母语者与英语双语使用者。本语料库中的麦克风语音均为英语，包含约3分钟与30分钟的访谈片段。

本补充数据集分为四个子部分，分别提供：
1. 分发给2008年SRE参与者的新增训练数据
2. 分发给2008年SRE后续评测参与者的追加数据
3. 2008年SRE主测试的访谈者信道文件（评测发布后公开）
4. 补充训练数据（评测发布后公开）

部分语音数据附带采用自动语音识别（Automatic Speech Recognition，简称ASR）系统生成的.cfm格式英语转写文本，但并非全部数据均包含此类转写。

若需查看本语料库包含的数据样例，请参阅该音频样例链接。

暂无最新更新。

部分内容 © 2007、2011 宾夕法尼亚大学理事会