Detecting Alarming Student Verbal Responses using Text and Audio Classifier

TL;DR

This paper introduces a hybrid text and audio classifier system to detect concerning student responses, improving safety and response review efficiency in automated verbal response scoring.

Contribution

A novel hybrid framework combining content and prosodic analysis for troubled student detection in AVRS systems.

Findings

Enhanced detection performance over traditional systems

Combines content and prosody for comprehensive analysis

Aims to facilitate quicker human review and intervention

Abstract

This paper addresses a critical safety gap in the use Automated Verbal Response Scoring (AVRS). We present a novel hybrid framework for troubled student detection that combines a text classifier, trained to detect responses based on their content, and an audio classifier, trained to detect responses using prosodic markers. This approach overcomes key limitations of traditional AVRS systems by considering both content and prosody of responses, achieving enhanced performance in identifying potentially concerning responses. This system can expedite the review process by humans, which can be life-saving particularly when timely intervention may be crucial.