Title :
The RWTH 2010 Quaero ASR evaluation system for English, French, and German
Author :
Sundermeyer, M. ; Nussbaum-Thom, M. ; Wiesler, S. ; Plahl, C. ; El-Desoky Mousa, A. ; Hahn, S. ; Nolden, D. ; Schlüter, R. ; Ney, H.
Author_Institution :
Comput. Sci. Dept., RWTH Aachen Univ., Aachen, Germany
Abstract :
Recognizing Broadcast Conversational (BC) speech data is a difficult task, which can be regarded as one of the major challenges beyond the recognition of Broadcast News (BN). This paper presents the automatic speech recognition systems developed by RWTH for the English, French, and German language which attained the best word error rates for English and German, and competitive results for the French task in the 2010 Quaero evaluation for BC and BN data. At the same time, the RWTH German system used the least amount of training data among all participants. Large reductions in word error rate were obtained by the incorporation of the new Bottleneck Multilayer Perception (MLP) features for all three languages. Additional improvements were obtained for the German system by applying a new language modeling technique, decomposing words into sublexical components.
Keywords :
broadcasting; information resources; multilayer perceptrons; natural language processing; performance evaluation; speech recognition; English language; French language; German language; MLP; RWTH 2010 Quaero ASR evaluation system; automatic speech recognition system; broadcast conversational speech data recognition; broadcast news recognition; multilayer perception; sublexical components; word error rates; Hidden Markov models; Mel frequency cepstral coefficient; Speech; Speech recognition; Training; Training data; automatic speech recognition; multilayer perceptrons;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
Conference_Location :
Prague
Print_ISBN :
978-1-4577-0538-0
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2011.5946920
