Non-native and accent speakers often face problems when using a speaker-independent (SI) speech recognition system. Speaker adaptation has been a solution to make SI recognizer work better for individuals. Targeting embedded implementation and applications in fast changing mobile environments, we in this paper proposed a supervised speaker adaptation (SA) solution with low system resource consumption, minimized disturbance to the data structure of SI recognizer, and superior adaptation performance. Adapted by UK speakers on a digit recognition task, the US English speech recognizer produced 65.9% digit error reduction. Other advantages of the proposed SA method include the multi-speaker adaptation, the fast adaptation, and the little changed speaker independency after adaptation.
Cite as: Zhang, Y., Wu, B., Ren, X., He, X. (2005) A speaker biased SI recognizer for embedded mobile applications. Proc. Interspeech 2005, 253-256, doi: 10.21437/Interspeech.2005-151
@inproceedings{zhang05b_interspeech, author={Yaxin Zhang and Bian Wu and Xiaolin Ren and Xin He}, title={{A speaker biased SI recognizer for embedded mobile applications}}, year=2005, booktitle={Proc. Interspeech 2005}, pages={253--256}, doi={10.21437/Interspeech.2005-151} }