Prosodic structure generation is the key component in improving the intelligibility and naturalness of synthetic speech for a text-to-speech (TTS) system. This paper investigates the problem of automatic segmentation of prosodic word and prosodic phrase,which are two fundamental layers in the hierarchical prosodic structure of Mandarin,and presents a two-stage prosodic structure generation strategy. Conditional random fields (CRF) models are built for both prosodic word and prosodic phrase prediction at the front end with diflerent feature selections. Besides,a transformation-based error-driven learning (TBL) modification module is introduced in the back end to amend the initial prediction. Experiment results show that the approach combining CRF and TBL achieves an F-score of 94.66%.
Pomo video recognition is important for Intemet content monitoring. In this paper, a novel pomo video recognition method by fusing the audio and video cues is proposed. Firstly, global color and texture features and local scale-invariant feature transform （SIFT） are extracted to train multiple support vector machine （SVM） classifiers for different erotic categories of image frames. And then, two continuous density hidden Markov models （CHMM） are built to recognize porno sounds. Finally, a fusion method based on Bayes rule is employed to combine the classification results by video and audio cues. The experimental results show that our model is better than six state-of-the-art methods.
A novel method for fingerprint singular points extraction including location and orientation is proposed based on some properties of the orientation field models. Singular points are located by clustering the results of corner detection. Then, through examining the sub-block orientation fields at a number of selected positions on concentric circles centered about the located singular point, an iterative method based on the orientation differences is proposed to compute the orientation of the core point. Experimental results on NIST4 and FVC2002 four databases demonstrate the proposed method can consistently locate singular points with the high accuracy. The location and orientation of the detected singular points can be used for alignment （translation and rotation） parameters in fingerprint matching.
