| Paper: | SLP-P1.3 | ||
| Session: | Feature Extraction and Modeling | ||
| Time: | Tuesday, May 16, 10:30 - 12:30 | ||
| Presentation: | Poster | ||
| Topic: | Speech and Spoken Language Processing: Feature Extraction and Modeling | ||
| Title: | Cross-domain and Cross-language Portability of Acoustic Features Estimated by Multilayer Perceptrons | ||
| Authors: | Andreas Stolcke, SRI International / University of California, Berkeley, United States; Frantisek Grezl, University of California, Berkeley, United States; Mei-Yuh Hwang, Xin Lei, University of Washington, United States; Nelson Morgan, University of California, Berkeley, United States; Dimitra Vergyri, SRI International, United States | ||
| Abstract: | Recent results with phone-posterior acoustic features estimated by multilayer perceptrons (MLPs) have shown that such features can effectively improve the accuracy of state-of-the-art large vocabulary speech recognition systems. MLP features are trained discriminatively to perform phone classification and are therefore, like acoustic models, tuned to a particular language and application domain. In this paper we investigate how portable such features are across domains and languages. We show that even without retraining, English-trained MLP features can provide a significant boost to recognition accuracy in new domains within the same language, as well as in entirely different languages such as Mandarin and Arabic. We also show the effectiveness of feature-level adaptation in porting MLP features to new domains. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012