The Community for Technology Leaders
Acoustics, Speech, and Signal Processing, IEEE International Conference on (2009)
Taipei, Taiwan
Apr. 19, 2009 to Apr. 24, 2009
ISBN: 978-1-4244-2353-8
pp: 4445-4448
Oriol Vinyals , International Computer Science Institute, Berkeley, CA, USA
Li Deng , Microsoft Research, Redmond, WA, USA
Dong Yu , Microsoft Research, Redmond, WA, USA
Alex Acero , Microsoft Research, Redmond, WA, USA
In this paper, 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.

O. Vinyals, Dong Yu, A. Acero and Li Deng, "Discriminative pronounciation learning using phonetic decoder and minimum-classification-error criterion," Acoustics, Speech, and Signal Processing, IEEE International Conference on(ICASSP), Taipei, Taiwan, 2009, pp. 4445-4448.
103 ms
(Ver 3.3 (11022016))