Detecting and analysing spontaneous oral cancer speech in the wild

TL;DR

This paper introduces a new dataset of spontaneous oral cancer speech from YouTube, analyzes key speech features for detection, and establishes baseline machine learning methods for identifying oral cancer speech in real-world conditions.

Contribution

It presents the first analysis of spontaneous oral cancer speech and sets baselines for detection using explainable machine learning techniques.

Findings

Sibilants and stop consonants are key indicators for detection.

Baseline models achieve measurable accuracy on the new dataset.

Spontaneous speech analysis differs from read speech in oral cancer detection.

Abstract

Oral cancer speech is a disease which impacts more than half a million people worldwide every year. Analysis of oral cancer speech has so far focused on read speech. In this paper, we 1) present and 2) analyse a three-hour long spontaneous oral cancer speech dataset collected from YouTube. 3) We set baselines for an oral cancer speech detection task on this dataset. The analysis of these explainable machine learning baselines shows that sibilants and stop consonants are the most important indicators for spontaneous oral cancer speech detection.