论文信息 - Simplified Simultaneous Perturbation Stochastic Approximation for the Optimization of Free Decoding Parameters

Simplified Simultaneous Perturbation Stochastic Approximation for the Optimization of Free Decoding Parameters

This paper deals with automatic optimization of free decoding parameters. We propose using a Simplified Simultaneous Perturbation Stochastic Approximation algorithm to optimize these parameters. This method provides a significant reduction in computational and labor costs. We also demonstrate that the proposed method successfully copes with the optimization of parameters for a specific target real-time factor, for all the databases we tested.

Aleksei Romanenko | Ivan Medennikov | Alexander Zatvornitsky

[1] J. Spall. Multivariate stochastic approximation using a simultaneous perturbation gradient approximation , 1992 .

[2] J. Kacur,et al. An Accuracy Optimization of a Dialog ASR System Utilizing Evolutional Strategies , 2007, 2007 5th International Symposium on Image and Signal Processing and Analysis.

[3] Thomas Hain,et al. Automatic Optimization of Speech Decoder Parameters , 2010, IEEE Signal Processing Letters.

[4] Brian Kan-Wing Mak,et al. Automatic estimation of decoding parameters using large-margin iterative linear programming , 2009, INTERSPEECH.

[5] Jochen Schwenninger,et al. Simultaneous perturbation stochastic approximation for automatic speech recognition , 2013, INTERSPEECH.

[6] J. Spall. Implementation of the simultaneous perturbation algorithm for stochastic optimization , 1998 .

[7] J. Kiefer,et al. Stochastic Estimation of the Maximum of a Regression Function , 1952 .

[8] Xiumei Yue. Improved Simultaneous Perturbation Stochastic Approximation and Its Application in Reinforcement Learning , 2008, 2008 International Conference on Computer Science and Software Engineering.