ITU-T P 85-1994 Method for Subjective Performance Assessment of the Quality of Speech Voice Output Devices - Telephone Transmission Quality Subjective Opinion Tests (Study Group 12.pdf
《ITU-T P 85-1994 Method for Subjective Performance Assessment of the Quality of Speech Voice Output Devices - Telephone Transmission Quality Subjective Opinion Tests (Study Group 12.pdf》由会员分享,可在线阅读,更多相关《ITU-T P 85-1994 Method for Subjective Performance Assessment of the Quality of Speech Voice Output Devices - Telephone Transmission Quality Subjective Opinion Tests (Study Group 12.pdf(13页珍藏版)》请在麦多课文档分享上搜索。
1、STD-ITU-T RECMN P-85-ENGL 1779 98b257L Ob177Lib 773 INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU TELEPHONE TRANSMISSION QUALITY SUBJECTIVE OPINION TESTS P.85 (06/94) A METHOD FOR SUBJECTIVE PERFORMANCE ASSESSMENT OF THE QUALITY OF SPEECH VOICE OUTPUT DE
2、VICES ITU-T Recommendation P.85 (Previously “CCIlT Recommendation”) - STD-ITU-T RECMN P.85-ENGL 199i W i8b257L Ob17747 b3T FOREWORD The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying tech
3、nical, operating and tariff questions and issuing Recommen- dations on them with a view to standardizing telecommunications on a worldwide basis. The World Telecornmunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups w
4、hich, in their turn. produce Recommendations on these topics. The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). ITU-T Recommendation P.85 was prepared by IT-T Study Group 12 (1993-1996) and was app
5、roved under the WTSC Resolution No. 1 procedure on the 21 of June 1994. NOTE In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. O ITU 1994 Ali rights reserved. No part of this publicat
6、ion may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. STD-ITU-T RECMN P-BCi-ENGL L77Li 48b257L DbL7948 57b M CONTENTS scope Assessment method 2.1 General 2.2 Main features of the reco
7、mmended method 3.1 Speech material . 3.2 Source conditions 3.3 Stimulus preparation . Design of experiment . 4.1 Subject task . 4.2 Rating scales . 4.3 Experimental design . 4.4 Listening test procedure Statistical analysis and reporting the results . Other Methods Test preparati on Annex A . Messag
8、es . Annex B - Response sheets . Annex C - Evaluation of synthetic speech: instructions for listeners . References . Bibliograph y Recommendation P.85 (06/94) Page 1 1 1 1 2 2 2 2 2 2 2 2 3 3 4 4 5 8 9 9 I STD-ITU-T RECMN P-85-ENGL 1774 W 48b2571 Ob19947 402 W SUMMARY Various services providing voca
9、l answers related to telephone directory inquiries, weather forecast, mail order, etc., are now available to PSTN users using vocal servers. As the speech messages are produced by machines, they may suffer from some impairment. In this Recommendation a method is defined for subjective performance as
10、sessment of the quality of speech of voice output devices. This method allows the comparison of several systems between them. It will be useful for system designers and service providers for checking the quality of their products. This method is of the listening test type. Messages are presented aur
11、ally to subjects. The subjects express their opinion on one or more rating scales after having answered specific questions on the information contained in the messages. The results are measures of the perceived quality in several aspects, which makes it possible to compare the effectiveness of diffe
12、rent speech synthesis systems. i Recommendation P.85 (04) STD-ITU-T RECMN P-85-ENGL 1774 48b2591 ObL9750 124 Recommendation P.85 A METHOD FOR SUBJECTIVE PERFORMANCE ASSESSMENT OF THE QUALITY OF SPEECH VOICE OUTPUT DEVICES (Geneva, 1994) 1 Scope Voice servers are now available for Public Switched Tel
13、ephone Network subscribers. These devices make use either of stored announcements or of synthetic speech. Synthetic speech may be produced from stored segments such as words, syllabes or diphones; it may also be produced by synthesis by rule, e.g. formant synthesis. In all cases of signal processing
14、, such as digital compression of the signal, together with sound processing such as concatenation of segments and variation of pitch, intensity and segment duration, a noticeable impairment of speech quality may occur. This Recommendation, based on Recommendation P.80 and specific experiments i, 2,
15、3, defines a testing method for evaluating the subjective quality of synthetic speech. Some adaptation of the method may be needed, depending on the type of system which is being evaluated. The method takes into account both the performance and the attitudes of the users. The attitudes are assessed
16、by the use of multiple scales. The Recommendation covers both overall system performance and the application to specific tasks. Two examples of application are provided in Annex A. This Recommendation is intended to describe a method for obtaining overall evaluations from users about the acoustic ou
17、tput of speech production devices. Procedures for evaluating specific components of text-to-speech systems (e.g. text transcription into phonetic units, etc.) are currently under study. 2 Assessment method 2.1 General The recommended methods for assessing telephony speech quality described in Recomm
18、endation P.80 and in 2.5 (Opinion tests) of the 2nd edition of Handbook on Telephonornerry 4 can be applied to the assessment of synthetic speech. The use of multiple opinion scales improves the description of listeners perception. Since synthetic speech may need some effort to be understood, the te
19、st is designed so that the subjects must pay attention to the information contained in messages before expressing their opinions. 2.2 Main features of the recommended method During a test a number of different voice sources will be presented aurally, so that the subjects opinions related to a given
20、source may be obtained in relation to other sources. The sources will be synthesis systems as well as reference conditions (this may include natural speech corrupted with some calibrated degradation and known synthesis systems). Subjects are asked to express their opinion using one or more Spint opi
21、nion scales, as in the Absolute Category Rating (ACR) or Degradation Category Rating (DCR) method of Recommendation P.80. In addition to the overall quality scale, other scales measuring listening effort, voice pleasantness, etc., can be used. The messages transmitted by the systems should be relate
22、d to practical applications. In practice different applications will require different test sessions. Each message is presented twice. During the first presentation subjects answer specific questions on the information contained in a message; then subjects judge the speech quality by expressing thei
23、r opinion on one or more rating scales during the second presentation. Recommendation P.85 (W94) 1 3 Test preparation 3.1 Speech material The messages should be long enough so that the subjects have time to reproduce the essential content on the first response sheet and also to give their opinion us
24、ing the rating scales on the second sheet. A duration of i0 to 30 seconds per message is recommended. Each message should consist of a fixed part which is specific to the task and a variable part which is different between pairs of presentation. The messages should be designed so that the predictabi
- 1.请仔细阅读文档,确保文档完整性,对于不预览、不比对内容而直接下载带来的问题本站不予受理。
- 2.下载的文档,不会出现我们的网址水印。
- 3、该文档所得收入(下载+内容+预览)归上传者、原创作者;如果您是本文档原作者,请点此认领!既往收益都归您。
本资源只提供5页预览,全部文档请下载后查看!喜欢就下载吧,查找使用更方便
10000 积分 0人已下载
下载 | 加入VIP,交流精品资源 |
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- ITUTP851994METHODFORSUBJECTIVEPERFORMANCEASSESSMENTOFTHEQUALITYOFSPEECHVOICEOUTPUTDEVICESTELEPHONETRANSMISSIONQUALITYSUBJECTIVEOPINIONTESTSSTUDYGROUP12PDF
![提示](http://www.mydoc123.com/images/bang_tan.gif)
链接地址:http://www.mydoc123.com/p-800679.html