首页 | 本学科首页   官方微博 | 高级检索  
     


Transputer application to speech recognition
Affiliation:1. Department of Computer Science, University of Hull, Hull HU6 7RX, UK;2. British Telecom Research Laboratories, Martlesham Heath, Ipswich IP5 7RE, UK;1. PG and Research Department of Physics, Queen Mary’s College (Autonomous), Chennai 600 004;2. Department of Physics, Stella Maris College (Autonomous), Chennai 600 086;1. Department of Nutrition, University of California Davis, Davis, CA, USA;2. Department of Community Health Sciences, College of Applied Medical Sciences, King Saud University, Riyadh, Saudi Arabia;3. Department of Internal Medicine, Division of Gastroenterology and Hepatology, University of California Davis, Sacramento, CA, USA;1. Dept of Electrical Engineering, COMSATS Institute of Information Technology, Islamabad, Pakistan;2. Internetworking Program, FE, Dalhousie University, Halifax, Canada;3. University of Alberta, Alberta, Canada;4. CAST, COMSATS Institute of Information Technology, Islamabad, Pakistan
Abstract:Use of the architectural features of transputers to provide realtime matching of speech recognition is considered. The dynamic time warping algorithm has been implemented in occam and tested initially for matching a single word against a single template. The implementation has been extended to the multitemplate case and preliminary results for this are given. Logical extension to a multitransputer system is also discussed. The single-transputer implementation has been undertaken using a T414 transputer resident in an IBM XT personal computer on an Inmos B004 evaluation board with 2 Mbyte of RAM.
Keywords:
本文献已被 ScienceDirect 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号