Abstract | ||
---|---|---|
Spotting recognition is the simultaneous realization of both recognition and segmentation. It is able to extract suitable information from an input dataset satisfying a query, and has developed into a research topic known as word spotting that uses dynamic programming or hidden Markov models. Continuous dynamic programming (CDP) is a promising method for spotting recognition applied to sequential patterns. However, the computational burden for conducting a retrieval task using CDP increases as O(JIP), where I is the input length, J is the reference length and P is the number of paths. This paper proposes a faster nonlinear spotting method like CDP, called fast spotter (FS). FS is regarded as an approximation of CDP using A* search. FS reduces the computational burden to O(IP log2 J) in the best case and executes in around half the time with an experimental dataset, enabling it to realize a large-scale speech retrieval system. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/CIT.2008.4594740 | CIT |
Keywords | Field | DocType |
large-scale speech retrieval system,spotting recognition,information retrieval,computational burden,fast spotter,sequential patterns,approximation algorithm,dynamic programming,hidden markov models,continuous dynamic programming,pattern recognition,speech recognition,approximation algorithms,speech,satisfiability,hidden markov model,databases,data mining | Dynamic programming,Distance measurement,Approximation algorithm,Speech retrieval,Nonlinear system,Pattern recognition,Computer science,Segmentation,Artificial intelligence,Hidden Markov model,Spotting | Conference |
ISBN | Citations | PageRank |
978-1-4244-2358-3 | 1 | 0.36 |
References | Authors | |
4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yuichi Yaguchi | 1 | 40 | 16.52 |
Keitaro Naruse | 2 | 47 | 19.98 |
Ryuichi Oka | 3 | 618 | 86.50 |