mgr inż. Szymon Zaporowski
English speech recognition experiments are presented employing both: audio signal and Facial Motion Capture (FMC) recordings. The principal aim of the study was to evaluate the inﬂuence of feature vector dimension reduction for the accuracy of vocalic segments classiﬁcation employing neural networks. Several parameter reduction strategies were adopted, namely: Extremely Randomized Trees, Principal Component Analysis and Recursive...
This paper provides a description and results of measurements of the maximum acceptable value of delay tolerated by a musician, while playing an instrument, that does not cause de-synchronization and discomfort. First, methodology of measurements comprising audio recording and a fast camera is described. Then, themeasurement procedure for acquiring the maximum value of delay conditioning...
This paper provides a description of experimental investigations concerning comparison between the audio path characteristics of various Android versions. First, information about the changes in each system version in the context of latency caused by them is presented. Then, a measurement procedure employing available applications to measure latency is described comparing to results contained in the Internet. Finally, a comparison...
wyświetlono 75 razy