engleski

Noise cancellation in automatic lip sync by speech signal analysis

This master thesis investigates noise cancellation techniques in order to enhance the performance of existing automatic lip synchronization system in noisy environments. Automatic lip synchronization is a method for generating virtual character face animation where animation is driven by the speech signal. The method consists of speech signal analysis, audio to visual mapping and face animation synthesis. Speech signal is mapped to visemes by speech feature recognition. The presence of background noise in the speech signal degrades the lip synchronization animation quality. The thesis investigates different speech signal enhancement methods, proposes and implements algorithms for noise cancellation as part of the enhanced automatic lip synchronization system. Proposed methods effectively track and cancel the noise in real time. To enhance speech signal aAdditional filtering is performed to aExisting speech recognition system performance is improved with the use of dynamic neural networks, resulting in overall better animation quality. Detailed system validation based on three different evaluation methods is done and potential applications of these technologies are discussed in details. This method is suitable for real-time and offline applications. It is speaker independent and multilingual.

Noise cancellation; lip sync

nije evidentirano