Novel LVCSR Decoder Based on Perfect Hash Automata and Tuple Structures – SPREAD –
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2014, Vol 5, Issue 5
Abstract
The paper presents the novel design of a one-pass large vocabulary continuous-speech recognition decoder engine, named SPREAD. The decoder is based on a time-synchronous beam-search approach, including statically expanded cross-word triphone contexts. An approach using efficient tuple structures is proposed for the construction of the complete search-network. The foremost benefits are the important space savings and higher processing speed, and the compact and reduced size of the tuple structure, especially when exploiting the structure of the key. In this way, the time needed to load the ASR search-network into the memory is also significantly reduced. Further, the paper proposes and presents the complete methodology for compiling general ASR knowledge sources into a tuple structures. Additionally, the beam search is enhanced with the novel implementation of a bigram language model Look-Ahead technique, by using tuple structures and a caching scheme. The SPREAD LVCSR decoder is based on a token-passing algorithm, capable of restricting its search-space by several types of token pruning. By using the presented language model Look-Ahead technique, it is possible to increase the number of tokens that can be pruned without decoding precision loss.
Authors and Affiliations
Matej Rojc, Kacic Zdravko
Keywords
Related Articles
- EP ID EP105025
- DOI 10.14569/IJACSA.2014.050504
- Views 97
- Downloads 0
Barriers in Adoption of Health Information Technology in Developing Societies
This paper develops the conceptual framework of barriers faced by the decision makers and management personnel of health sector. The main theme of this paper is to give a clear understanding about the adaption barriers o...
A Distributed Approach based on Transition Graph for Resolving Multimodal Urban Transportation Problem
All over the world, many research studies focus on developing and enhancing real-time communications between various transport stakeholders in urban environments. Such motivation can be justified by the growing importanc...
Performance Comparison of QEC Network based JAVA Application and Web based PHP Application
Every organization wants to automate the manual system for moving and storing their data in particular format. A QEC department takes feedback of teacher evaluation manually from the students in the university that is so...
Impact of Domain Modeling Techniques on the Quality of Domain Model: An Experiment
The unified modeling language (UML) is widely used to analyze and design different software development artifacts in an object oriented development. Domain model is a significant artifact that models the problem domain a...
Rule-based Emotion AI in Arabic Customer Review
The e-commerce emotion analysis is notable and the most pivotal advance since it catches the customer emotion in a product, and emotions with respect to product to decide if the customer attitude is negative, positive, o...