Font Size: a A A

Research Of Intelligent HCI Technology Based On Speech Recogniton

Posted on:2013-05-23Degree:MasterType:Thesis
Country:ChinaCandidate:L ChenFull Text:PDF
GTID:2248330371481358Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
With the maturity of speech recognition theory, some related algorithm in The practical process of this technology has been adopted. Speech recognition technology is developing further toward the practical direction. Human-computer interaction (HCI) technology is a study that is mutual understanding and communication (including input and output) between people and computer, to help people complete a technology of information management services and processing functions. A new generation of human-computer interaction technology is a hot topic of the last20years in the field of computer science, and gets great deal of attention at home and abroad. Research institutions and companies have invested substantial human and financial resources for innovative and practical research, in which intelligent human-computer interaction technology is a key technology. This is verified in the actual application:intelligent electronic products such as smart phones and Tablet PC competition have evolved to become the competition of operating experience. The new generation of human-computer interaction emphasis on the following characteristics:the "people-centered", the multi-modal interaction, intelligent perception, multi-dimensional environment interaction and so on. A data from the second half of2011shows that, smart phones and Tablet PCs have more shipments than the PC for the first time. This hinted that the Earth has entered the mobile Internet era and the handheld computing has become one of today’s most important computing model. But the low naturalness of the HCI and the efficiency of interactive exposure obviously, multi-channel interactive and multi-channel user interface can effectively improve the efficiency of its human-computer interaction. Input method based on speech recognition and speech synthesis output method interactively integrate into the mode of a new generation of human-computer interaction has become a new way to improve the means of human-computer interaction, having the necessity of explore and the higher actual value.The article analyses the key link of the voice intelligent HCI technology achievement, namely the establishment of speech model of the input channel, including advantages and defects of the speech channels, and introduces the corresponding solutions. The article also introduces some other auxiliary interactive input modes. The focus analysis of the article is the methods of the speech signal identification, including the generation model of the speech signal, pre-processing, parameter extraction and processing, and recognition algorithms. Then, it analyzes the mapping from speech library of the digital music players command to the control command. Finally, it analyzes the establishment of a hand-held mobile platform HCI intelligent speech input channel, and the fulfillment on the popular mobile intelligent operation system Android. The mobile terminal at all times has a networking advantage. It explores the technologies, which use the server to complete the identification and to reduce the mobile terminal computing burden, and gives the specific software development content. The entire article content covers the building process of input channels of speech recognition intelligence HCI.It analyzes the pre-emphasis and endpoint detection in the speech front-end processing, LPC, LPCC Features and MFCC parameters in the choice of speech parameters, and the DTW-based template matching algorithm, combining with the processing capabilities and the practical application demands of the recent mobile terminal in the speech recognition model, as well as the MKM clustering algorithm. Through the simulation analysis can made some meaningful results. The algorithm can guarantee real-time requirements to obtain a higher recognition rate, which for practical application to provide a guarantee. This project analyzes and evaluates interactive needs, and analyzes speech interaction model, facing the application of human-computer interaction in the digital music player platform and targeting at the actual needs of the platform in mobile environment. With regard to specific software design demands of hand-held computing platform Android, its realization of intelligent HCI system achieves a friendly human-computer natural interaction.
Keywords/Search Tags:Speech recognition, Intelligent HCI, Speech synthesis, SVM, Handheldcomputing, Android
PDF Full Text Request
Related items