Confidence of agreement among multiple LVCSR models and model combination by SVM

For many practical applications of speech recognition systems, it is quite desirable to have an estimate of confidence for each hypothesized word. Unlike previous works on confidence measures, we have proposed features for confidence measures that are extracted from outputs of more than one LVCSR models. For further analysis of the proposed confidence measure, this paper examines the correlation between each word´s confidence and the word´s features such as its part-of-speech and syllable length. We then apply SVM learning technique to the task of combining outputs of multiple LVCSR models, where, as features of SVM learning, information such as the pairs of the models which output the hypothesized word are useful for improving the word recognition rate. Experimental results show that the combination results achieve a relative word error reduction of up to 72 % against the best performing single model and that of up to 36 % against ROVER.