%Aigaion2 BibTeX export from Idiap Publications
%Sunday 14 April 2024 06:55:37 AM

         author = {VILLATORO-TELLO, Esa{\'{u}} and Madikeri, Srikanth and Sharma, Bidisha and Khalil, Driss and Kumar, Shashi and Iuliia, Nigmatulina and Motlicek, Petr and Ganapathiraju, Aravind},
       keywords = {Cross-modal Alignment, Intent Classification, knowledge distillation, Spoken Language Understanding, Word-Confusion-Networks},
       projects = {UNIPHORE},
          month = apr,
          title = {Probability-Aware Word-Confusion-Network-to-Text Alignment Approach for Intent Classification},
      booktitle = {Proceedings of the 49th IEEE International Conference on Acoustics, Speech, & Signal Processing (ICASSP) 2024},
           year = {2024},
       abstract = {Spoken Language Understanding (SLU) technologies have seen a big improvement due to the effective pretraining of speech representations. A common requirement of industry-based solutions is the portability to deploy SLU models in voice-assistant devices. Thus, distilling knowledge from large text-based language models has become an attractive solution for achieving good performance and guaranteeing portability. In this paper, we introduce a novel architecture that uses a cross-modal attention mechanism to extract bin-level contextual embeddings from a word-confusion network (WNC) encoding such that these can be directly compared and aligned with traditional text-based contextual embeddings. This alignment is achieved using a recently proposed tokenwise constrastive loss function. We validated our architecture's effectiveness by fine-tuning our WCN-based pretrained model to perform intent classification on the SLURP dataset. Obtained accuracy (81\%), depicts a 9.4\% relative improvement compared to a recent and equivalent E2E method.}