Dialogue history integration into end-to-end signal-to-concept spoken language understanding systems
This work addresses the challenge of leveraging context in SLU systems for better accuracy, but it is incremental as it builds on existing end-to-end methods with specific embedding types.
The paper tackled the problem of integrating dialogue history into end-to-end spoken language understanding systems to improve performance, and the result was that proposed dialog history embeddings (h-vectors) enhanced model accuracy on the MEDIA corpus for semantic slot filling.
This work investigates the embeddings for representing dialog history in spoken language understanding (SLU) systems. We focus on the scenario when the semantic information is extracted directly from the speech signal by means of a single end-to-end neural network model. We proposed to integrate dialogue history into an end-to-end signal-to-concept SLU system. The dialog history is represented in the form of dialog history embedding vectors (so-called h-vectors) and is provided as an additional information to end-to-end SLU models in order to improve the system performance. Three following types of h-vectors are proposed and experimentally evaluated in this paper: (1) supervised-all embeddings predicting bag-of-concepts expected in the answer of the user from the last dialog system response; (2) supervised-freq embeddings focusing on predicting only a selected set of semantic concept (corresponding to the most frequent errors in our experiments); and (3) unsupervised embeddings. Experiments on the MEDIA corpus for the semantic slot filling task demonstrate that the proposed h-vectors improve the model performance.