Noise-Agnostic Multitask Whisper Training for Reducing False Alarm Errors in Call-for-Help Detection

TL;DR

This paper introduces a noise-agnostic multitask learning method for call-for-help detection that leverages pretrained ASR models to reduce false alarms caused by environmental noise, improving robustness and scalability.

Contribution

The paper proposes a novel noise-agnostic multitask training approach that adds a noise classification head to pretrained ASR models, enhancing robustness in real-world call-for-help detection scenarios.

Findings

Significant reduction in false alarms in noisy environments

Improved robustness of call-for-help detection systems

Efficient multitask learning approach

Abstract

Keyword spotting is often implemented by keyword classifier to the encoder in acoustic models, enabling the classification of predefined or open vocabulary keywords. Although keyword spotting is a crucial task in various applications and can be extended to call-for-help detection in emergencies, however, the previous method often suffers from scalability limitations due to retraining required to introduce new keywords or adapt to changing contexts. We explore a simple yet effective approach that leverages off-the-shelf pretrained ASR models to address these challenges, especially in call-for-help detection scenarios. Furthermore, we observed a substantial increase in false alarms when deploying call-for-help detection system in real-world scenarios due to noise introduced by microphones or different environments. To address this, we propose a novel noise-agnostic multitask learning approach that integrates a noise classification head into the ASR encoder. Our method enhances the model's robustness to noisy environments, leading to a significant reduction in false alarms and improved overall call-for-help performance. Despite the added complexity of multitask learning, our approach is computationally efficient and provides a promising solution for call-for-help detection in real-world scenarios.