Parallel scalability in speech recognition
IEEE Signal Processing Magazine2009Vol. 26(6), pp. 124–135
Citations Over TimeTop 10% of 2009 papers
Kisun You, Jike Chong, Youngmin Yi, Ekaterina Gonina, Christopher J. Hughes, Yen-Kuang Chen, Wonyong Sung, Kurt Keutzer
Abstract
Parallel scalability allows an application to efficiently utilize an increasing number of processing elements. In this article, we explore a design space for parallel scalability for an inference engine in large vocabulary continuous speech recognition (LVCSR). Our implementation of the inference engine involves a parallel graph traversal through an irregular graph-based knowledge network with millions of states and arcs. The challenge is not only to define a software architecture that exposes sufficient fine-grained application concurrency but also to efficiently synchronize between an increasing number of concurrent tasks and to effectively utilize parallelism opportunities in today's highly parallel processors.
Related Papers
- → Basic Definitions and Network Traversal Algorithms(2013)8 cited
- → Exploring graph traversal algorithms in graph-based molecular generation(2021)1 cited
- Discovery of Frequent Traversal Patterns from Weighted Traversals and Performance Enhancement by Traversal Split(2007)
- → Exploring graph traversal algorithms in graph-based molecular generation(2021)
- → Exploring graph traversal algorithms in graph-based molecular generation(2021)