Event

Discriminative Pronunciation Learning Using Phonetic Decoder and Minimum-Classification-Error Criterion

Oriol Vinyals

ICSI

Tuesday, October 27, 2009
12:30

In this this talk, which summarizes the work I did as a summer intern at MSR under Li Deng's supervision, we report our recent research aimed at improving the pronunciation-modeling component of a speech recognition system designed for mobile voice search. Our new discriminative learning technique overcomes the limitation of the traditional ways of introducing alternative pronunciations that often enlarge confusability across different lexical items. Instead, we make use of a phonetic recognizer to generate pronunciation candidates, which are then evaluated and selected using the global minimum-classification error measure, guaranteeing a reduction of the training-set error rate after introducing alternative pronunciations. A maximum entropy approach is subsequently used to learn the weight parameters of the selected pronunciation candidates. Our experimental results demonstrate the effectiveness of the discriminative pronunciation learning technique in a real-world speech recognition task where pronunciation of business names presents special difficulty for high-accuracy speech recognition.