In this paper we study various front-end features, modeling and adaptation algorithms on the Aurora 3 databases, including auditory, moment, and AM-FM modulation features, context-dependent digit models, segmental K-means training, discriminative training, and model adaptations. The evaluation results on Aurora 3 are presented with a brief summary of our Aurora 2 results. Evaluation of Robust Speech Recognition
Cite as: Chen, J., Dimitriadis, D., Jiang, H., Li, Q., Myrvoll, T.A., Siohan, O., Soong, F.K. (2002) Bell labs approach to Aurora evaluation on connected digit recognition. Proc. 7th International Conference on Spoken Language Processing (ICSLP 2002), 229-232, doi: 10.21437/ICSLP.2002-9
@inproceedings{chen02_icslp, author={Jingdong Chen and Dimitris Dimitriadis and Hui Jiang and Qi Li and Tor André Myrvoll and Olivier Siohan and Frank K. Soong}, title={{Bell labs approach to Aurora evaluation on connected digit recognition}}, year=2002, booktitle={Proc. 7th International Conference on Spoken Language Processing (ICSLP 2002)}, pages={229--232}, doi={10.21437/ICSLP.2002-9} }