We build a two-pass Mandarin Automatic Speech Recognition (ASR) decoder on mobile device (PDA). The first-pass recognizing base syllable is implemented by discrete Hidden Markov Model (HMM) with time-synchronous, tree-lexicon Viterbi search. The second-pass dealing with language model, pronunciation lexicon and N-best syllable hypotheses from first-pass is implemented by Weighted Finite State Transducer (WFST). The best word sequence is obtained by shortest path algorithms over the composition result. This system limits the application in travel domain and it decouples the application of acoustic model and the application of language model into independent recognition passes. We report the real-time recognition performance performed on ASUS P565 with a 800MHz processor, 128MB RAM running Microsoft Window Mobile 6 operating system. The 26-hour TCC-300 speech data is used to train 151 acoustic model. The 3-minute speech data recorded by reading the travel-domain transcriptions is used as the testing set for evaluating the performances (syllable, character accuracies) and real-time factors on PC and on PDA. The trained bi-gram model with 3500-word from BTEC corpus is used in second-pass. In the first-pass, the best syllable accuracy is 38.8% given 30-best syllable hypotheses using continuous HMM and 26-dimension feature. Under the above syllable hypotheses and acoustic model, we obtain 27.6% character accuracy on PC after the second-pass.

ITC/USA 2013 Conference Proceedings / The Forty-Ninth Annual International Telemetering Conference and Technical Exhibition / October 21-24, 2013 / Bally's Hotel & Convention Center, Las Vegas, NV / The transition of telemetry from link-based to network-based architectures opens these systems to new security risks. Tools such as intrusion detection systems and vulnerability scanners will be required for emerging telemetry networks. Intrusion detection systems protect networks against attacks that occur once the network boundary has been breached. An intrusion detection model was developed in the Wireless Networking and Security lab at Morgan State University. The model depends on network traffic being filtered into traffic streams. The streams are then reduced to vectors. The current state of the network can be determined using Viterbi analysis of the stream vectors. Viterbi uses the output of the Hidden Markov Model to find the current state of the network. The state information describes the probability of the network being in predefined normal or attack states based on training data. This output can be sent to a network administrator depending on threshold levels. In this project, a penetration-testing tool called Metasploit was used to launch attacks against systems in an isolated test bed. The network traffic generated during an attack was analyzed for use in the MSU intrusion detection model.

No description available.

In telecommunication, the series of mobile network exceptions is a processwhich exhibits surges and bursts. The bursty part is usually caused by systemmalfunction. Additionally, the mobile network exceptions are often timedependent. A model that successfully captures these aspects will make troubleshootingmuch easier for system engineers. The Hidden Markov Model(HMM) is a good candidate as it provides a mechanism to capture both thetime dependency and the random occurrence of bursts. This thesis focuses onan application of the HMM to mobile network exceptions, with a case study ofEricsson’s Dropped Call data. For estimation purposes, two methods of maximumlikelihood estimation for HMM, namely, EM algorithm and stochasticEM algorithm, are used.

This thesis deals with scene level summarization of story-based videos. Two different approaches for story-based video summarization are investigated. The first approach probabilistically models the input video and identifies scene boundaries using the same model. The second approach models scenes and classifies scene types by evaluating likelihood values of these models. In both approaches, hidden Markov models are used as the probabilistic modeling tools. The first approach also exploits the relationship between video summarization and video production, which is briefly explained, by means of content types. Two content types are defined, dialog driven and action driven content, and the need to define such content types is emonstrated by simulations. Different content types use different hidden Markov models and features. The selected model segments input video as a whole. The second approach models scene types. Two types, dialog scene and action scene, are defined with different features and models. The system classifies fixed sized partitions of the video as either of the two scene types, and segments partitions separately according to their scene types. Performance of these two systems are compared against a iv deterministic video summarization method employing clustering based on visual properties and video structure related rules. Hidden Markov model based video summarization using content types enjoys the highest performance.

Zugl.: Basel, Univ., Diss., 2007

Thesis (M.Eng.(Computer Engineering)--University of Pretoria, 2008. / Summaries in Afrikaans and English. Includes bibliographical references.

Zugl.: München, Universiẗat der Bundeswehr, Diss., 2009.

Thesis (Ph.D)--Electrical and Computer Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Lee, Chin-Hui; Committee Member: Anthony Joseph Yezzi; Committee Member: Biing-Hwang (Fred) Juang; Committee Member: Mark Clements; Committee Member: Ming Yuan. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Thesis (Ph.D)--Electrical and Computer Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Dr. Chin-Hui Lee; Committee Member: Dr. Anthony Joseph Yezzi; Committee Member: Dr. Biing-Hwang (Fred) Juang; Committee Member: Dr. Mark Clements; Committee Member: Dr. Ming Yuan. Part of the SMARTech Electronic Thesis and Dissertation Collection.