Text-dependent speaker verification is becoming popular in the speaker recognition society. However, the conventional i-vector framework which has been successful for speaker identification and other similar tasks works relatively poorly in this task. Researchers have proposed several new methods to improve performance, but it is still unclear that which model is the best choice, especially when the pass-phrases are prompted during enrollment and test. In this paper, we introduce four modeling methods and compare their performance on the newly published RedDots dataset. To further explore the influence of different frame alignments, Viterbi and forwardbackward algorithms are both used in the HMM-based models. Several bottleneck features are also investigated. Our experiments show that, by explicitly modeling the lexical content, the HMMbased modeling achieves good results in the fixed-phrase condition. In the prompted-phrase condition, GMM-HMM and i-vector/HMM are not as successful. In both conditions, the forward-backward algorithm brings more benefits to the i-vector/HMM system. Additionally, we also find that even though bottleneck features perform well for text-independent speaker verification, they do not outperform MFCCs on the most challenging Imposter-Correct trials on RedDots.
In this work, we present an ultrasonic actuator that can work under cryogenic vacuum environment. It can be used for adjusting distance between capacitor electrodes in high temperature superconductor filter (HTSF) to tune its pass-band. The actuator is an single crystal chips driven nut-type ultrasonic motor, which can work under cryogenic vacuum conditions. The stator of a nut-type ultrasonic motor is a nut-shaped octagon with internal thread, which matches with the rotor external thread and a bottom at one end as fixing base. Piezoelectric chips are glued to the sides of the octagon to generate a traveling wave along the circumference. Vibration of the stator drives the rotor to rotate via friction between thread interfaces and the thread drives the rotor to move along the axis direction. The actuator that we developed can easily acquire micrometer positioning accuracy, which enables it to tune the pass-band of a HTSF effectively. The motor was optimized with FEM harmonic response analysis. The mechanical characteristics and stepping precision of the prototype ultrasonic motor have been measured and discussed. [Work supported by NSFC.]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.