Minimizing Computational Cost for Dynamic Programming Algorithms
Author | : Alex Waibel |
Publisher | : |
Total Pages | : 21 |
Release | : 1981 |
ISBN-10 | : OCLC:227502872 |
ISBN-13 | : |
Rating | : 4/5 ( Downloads) |
Download or read book Minimizing Computational Cost for Dynamic Programming Algorithms written by Alex Waibel and published by . This book was released on 1981 with total page 21 pages. Available in PDF, EPUB and Kindle. Book excerpt: In this study we introduce and test several methods to reduce the computational cost in dynamic programming algorithms for isolated word recognition systems. Three methods will be discussed in detail: (1) Pruning by preset thresholds, (2) Search based on the Branch and Bound technique, (3) Branch and Bound based search with additional pruning. Compared to conventional algorithms, Method 3 could be seen to yield a speed up of approximately a factor of 5, at no loss of recognition accuracy. The branch and bound method with pruning is also ideally suited for research oriented systems, since pruning is independent of the parametrization used (eliminates the necessity for retuning thresholds). Additional features of this method, which are of importance to maintaining the flexibility and diagnosticity needed for such a system, will be discussed. (Author).