12th Annual Conference of the International Speech Communication Association

Florence, Italy
August 27-31. 2011

AUC Optimization Based Confidence Measure for Keyword Spotting

Haiyang Li, Jiqing Han, Tieran Zheng

Harbin Institute of Technology, China

Confidence measure plays an important role in keyword spotting. To enhance the effectiveness of the confidence measure, we propose a novel method which improves the performance of keyword spotting by directly maximizing the area under the ROC curve (AUC). Firstly, we approximate the AUC as an objective function with the weighted mean confidence measure. Then, we optimize the objective function by training the weighting factors with the generalized probabilistic descent algorithm. Compared with the current method based on minimum classification error (MCE) criterion, the proposed method makes a global enhancement of ROC curve and does not need to train any threshold. The experiments conducted on the King-ASR-023 database show that the proposed method outperforms both the method averaging phone-level confidences and the method based on MCE.

Full Paper

Bibliographic reference.  Li, Haiyang / Han, Jiqing / Zheng, Tieran (2011): "AUC optimization based confidence measure for keyword spotting", In INTERSPEECH-2011, 1917-1920.