Spelling suggestions: "subject:"forminformation 1heory,"" "subject:"forminformation btheory,""
171 |
Separability of finite discrete signalsTamvaclis, C. January 1982 (has links)
No description available.
|
172 |
Modelling and classifying speech patternsNiranjan, Mahesan January 1990 (has links)
No description available.
|
173 |
Discriminative methods in HMM-based speech recognitionValtchev, Valtcho January 1995 (has links)
No description available.
|
174 |
Speech processing with neural networksWu, Lizhong January 1992 (has links)
No description available.
|
175 |
Real-time digital image processing in electron microscopyErasmus, S. J. January 1982 (has links)
No description available.
|
176 |
Multipulse excited linear predictive analysis in speech coding and constructive speech synthesisVarga, A. P. January 1985 (has links)
No description available.
|
177 |
Improved speech hidden Markov modelling via an expectation-maximization frameworkWong, Georges January 1992 (has links)
No description available.
|
178 |
Microprocessor-based controllers for induction motorsWijesundera, Dhammika Sarath January 1989 (has links)
No description available.
|
179 |
Error detection abilities of conducting students under four modes of instrumental score study.Crowe, Don Raymond. January 1994 (has links)
This study investigated the effect of four score study styles--no score study, study with score alone, study with score and a correct aural example, and score study at the electronic keyboard--on the pitch and rhythm error detection abilities of beginning conducting students. Subjects were 30 members of undergraduate beginning conducting classes at three midwestern universities. Four tests were developed, each having 31 four- to six-measure excerpts from band literature. Each excerpt contained only one error. Excerpts were grouped according to difficulty and assigned to tests in a modified random manner to facilitate equality of difficulty between sets. Within each test, excerpts were arranged in order of increasing difficulty and rescored to contain from one to eight parts. A counterbalanced design was utilized featuring a Latin Square into which the four score study styles were entered. Over the course of four sessions subjects received all four styles and all four tests. The orders in which subjects received score study styles were assigned on a rotational basis. Each subject within a university received the tests in the same order, but this order varied between universities. Six Hypercard © (Atkinson, 1987-90) stacks were developed on a Macintosh LC computer for presentation of the tests, management of the study, and data collection. Excerpts were played through MIDI keyboards using sampled wind instrument sounds. Study with the score and a correct aural example was found to be significantly more effective than either study with the score alone or no study. No significant difference was found between score study at the keyboard and any other score study style. There were significant differences in test scores attributable to the number of parts in examples. Generally, error detection became more difficult as the number of parts in examples increased. There were no significant differences in test scores attributable to the order of presentation of score study styles, individual example sets, or groups/order of presentation of example sets. There were significant differences in means score study time per session attributable to score study style, and in mean total time per session attributable to session number.
|
180 |
Signal processing techniques for partial discharge detection and mapping in high voltage underground cable networksShim, Ivy Ai Fei January 2001 (has links)
No description available.
|
Page generated in 0.1069 seconds