This thesis explores the use of semantic parsing for improving speech recognition performance. Specifically, it explores how a semantic parser may be used in order to re-rank the n-best hypothesis list generated by an automatic speech recognition system. We also explore how system performance is affected when retraining the system's acoustic model using a portion of the re-ranked data.

PDF

Undergraduate Honors Thesis, Computer Science Department, University of Texas at Austin.

@misc{corona:thesis16, title={An Analysis of Using Semantic Parsing for Speech Recognition}, author={Rodolfo Corona}, note={Undergraduate Honors Thesis, Computer Science Department, University of Texas at Austin}, url="http://www.cs.utexas.edu/users/ai-labpub-view.php?PubID=127609", year={2016} }

Slides (PPT)