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 116
- Downloads 0
Secure Data Accumulation among Reliable Hops with Rest/Alert Scheduling in Wireless Sensor Networks
Wireless Sensor Networks (WSNs) are more inclined to attackers by outer sources. The total information must be secured to guarantee the uprightness and privacy. In sensor networks, the data collection and data accumulati...
Deep Learning Classification of Biomedical Text using Convolutional Neural Network
In this digital era, the document entries have been increasing days by days, causing a situation where the volume of the document entries in overwhelming. This situation has caused people to encounter with problems such...
The Role of Camera Convergence in Stereoscopic Video See-through Augmented Reality Displays
In the realm of wearable augmented reality (AR) systems, stereoscopic video see-through displays raise issues related to the user’s perception of the three-dimensional space. This paper seeks to put forward few considera...
Feature Fusion for Negation Scope Detection in Sentiment Analysis: Comprehensive Analysis over Social Media
Negation control for sentiment analysis is essential and effective decision support system. Negation control include identification of negation cues, scope of negation and their influence within it. Negation can either s...
Phishing Website Detection based on Supervised Machine Learning with Wrapper Features Selection
The problem of Web phishing attacks has grown considerably in recent years and phishing is considered as one of the most dangerous Web crimes, which may cause tremendous and negative effects on online business. In a Web...