A Convolutional Neural Network for Search Term Detection

TL;DR

This paper introduces Halo, an indoor hospital navigation app that uses a custom CNN to accurately detect search terms from speech transcriptions, improving guidance without relying on localization systems.

Contribution

It presents a novel CNN model specifically trained for hospital-related search term detection from speech transcriptions, outperforming traditional matching methods.

Findings

CNN achieves higher accuracy than Levenshtein distance matching.

Model tailored for hospital environment improves search term detection.

Enhanced indoor navigation without localization systems.

Abstract

Pathfinding in hospitals is challenging for patients, visitors, and even employees. Many people have experienced getting lost due to lack of clear guidance, large footprint of hospitals, and confusing array of hospital wings. In this paper, we propose Halo; An indoor navigation application based on voice-user interaction to help provide directions for users without assistance of a localization system. The main challenge is accurate detection of origin and destination search terms. A custom convolutional neural network (CNN) is proposed to detect origin and destination search terms from transcription of a submitted speech query. The CNN is trained based on a set of queries tailored specifically for hospital and clinic environments. Performance of the proposed model is studied and compared with Levenshtein distance-based word matching.