SpeakNav

Many navigation applications take natural language speech as input, which avoids users typing in words and thus improves traffic safety. However, often fail to understand a user's free-form description of route. In addition, they only support input specific source or destination, does not enable specify additional route requirements. We propose SpeakNav framework that enables describe intended routes via then recommends appropriate routes. Specifically, we novel Route Template based Bidirectional Encoder Representation from Transformers (RT-BERT) model supports the understanding descriptions. The extraction information POI keywords related distances. Then formalize template-driven path query uses extracted information. To efficient processing, develop hybrid label index for computing network distances between POIs, branch-and-bound algorithm along with pivot reverse B-tree (PB-tree) index. Experiments real synthetic data indicate RT-BERT offers high accuracy proposed is capable outperforming baseline algorithms.