Font Size: a A A

The Research Of Speech Recognition Technology Based On The Wearable Computing Platform

Posted on:2008-07-09Degree:MasterType:Thesis
Country:ChinaCandidate:L M ZhuFull Text:PDF
GTID:2178360215964675Subject:Computer system architecture
Abstract/Summary:PDF Full Text Request
Human beings have entered the post PC era and pervasive computing has become an important growing aspect. The essence is to create an environment that connects people's life seamlessly and has the ability of computing and communicating. As the reasons above, wearable computers appear. With all kinds of wearable computing equipment and mobile computing equipment increase unceasingly, the requirements towards them become higher and higher. The requirements include making their bulk as small as possible, the power as low as possible and carrying on information input everywhere at any time. So the traditional keyboard input manner can't satisfy the requirement that inputs information conveniently and naturally. Speech recognition technology can change human-computer interaction means effectively and play an important role in wearable computing input technology.The content of the research subject comes from Special Science and Research Plan Project of Shaanxi education office: the research on wearable computing platform technology. The author of this dissertation takes charge in the research and development of speech recognition module based on wearable computing platform in this project. The dissertation is based on this project.According to the subject requirement and the comparison with other people's work done before, an embedded speech recognition system based on special person and isolated words should be implemented. As the emphasis and difficulty of the dissertation, the system's overall design, the hardware platform design and the arithmetic of some important sub modules are illustrated. Particularly, endpoint detection module, characteristic distilled module and pattern matching module are analyzed and improved. The transplant of the software onto the embedded platform is also introduced.It gives an evaluation and analysis result of the system in the last part of the dissertation. The recognition precision can reach 89.5%. The advantages and disadvantages of the system are also summarized and the prospect is put forward. The research content of the dissertation will promote the further research on wearable computing platform technology.
Keywords/Search Tags:embedded system, wearable computer platform, speech recognition technology
PDF Full Text Request
Related items