Integrating multi-level linguistic knowledge with a unified framework for Mandarin speech recognition

Abstract

To improve the Mandarin large vocabulary continuous speech recognition (LVCSR), a unified framework based approach is introduced to exploit multi-level linguistic knowledge. In this framework, each knowledge source is represented by a Weighted Finite State Transducer (WFST), and then they are combined to obtain a so-called analyzer for integrating multi-level knowledge sources. Due to the uniform transducer representation, any knowledge source can be easily integrated into the analyzer, as long as it can be encoded into WFSTs. Moreover, as the knowledge in each level is modeled independently and the combination is processed in the model level, the information inherently in each knowledge source has a chance to be thoroughly exploited. By simulations, the effectiveness of the analyzer is investigated, and then a LVCSR system embedding the presented analyzer is evaluated. Experimental results reveal that this unified framework is an effective approach which significantly improves the performance of speech recognition with a 9.9% relative reduction of character error rate on the HUB-4 test set, a widely used Mandarin speech recognition task.

Prosodic modeling with rich syntactic context in HMM-based Mandarin speech synthesis

“…In these methods, a statistical model is firstly trained by extracting lexical and syntactic features and then applied to predict prosodic structure of unlabel texts. Finally, prosodic structure is determined mainly relying on syntactic information and then used to model prosodic parameters [6].…”