Syntactic Reanalysis in Language Models for Speech Recognition

State-of-the-art speech recognition systems steadily increase their performance using different variants of deep neural networks and postprocess the results by employing N-gram statistical models trained on a large amount of data coming from the general-purpose domain. While achieving an excellent performance regarding Word Error Rate (17.343% on our HumanRobot Interaction data set), state-of-the-art systems generate hypotheses that are grammatically incorrect in 57.316% of the cases. Moreover, if employed in a restricted domain (e.g. HumanRobot Interaction), around 50% of the hypotheses contain out-ofdomain words. The latter are confused with similarly pronounced in-domain words and cannot be interpreted by a domain-specific inference system. The state-of-the-art speech recognition systems lack a mechanism that addresses the syntactic correctness of hypotheses. We propose a system that can detect and repair grammatically incorrect or infrequent sentence forms. It is inspired by a computational neuroscience model that we developed previously. The current system is still a proof-of-concept version of a future neurobiologically more plausible neural network model. Hence, the resulting system postprocesses sentence hypotheses of state-ofthe-art speech recognition systems, producing in-domain words in 100% of the cases, syntactically and grammatically correct hypotheses in 90.319% of the cases. Moreover, it reduces the Word Error Rate to 11.038%.

@InProceedings{THW17, 
 	 author =  {Twiefel, Johannes and Hinaut, Xavier and Wermter, Stefan},  
 	 title = {Syntactic Reanalysis in Language Models for Speech Recognition}, 
 	 booktitle = {Proceedings of the 7th Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EpiRob)},
 	 editors = {},
 	 number = {},
 	 volume = {},
 	 pages = {215--220},
 	 year = {2017},
 	 month = {Sep},
 	 publisher = {IEEE},
 	 doi = {10.1109/DEVLRN.2017.8329810}, 
 }