Few-Shot Contrastive Adaptation for Audio Abuse Detection in Low-Resource Indic Languages

TL;DR

This paper explores the use of contrastive language-audio pre-training (CLAP) for detecting abusive speech directly from audio in low-resource, multilingual settings, showing promising cross-lingual representations and adaptation strategies.

Contribution

It demonstrates that CLAP-based audio representations can support effective cross-lingual abuse detection with minimal supervised data, highlighting the potential of contrastive audio-text models in low-resource languages.

Findings

CLAP yields strong cross-lingual audio representations across ten Indic languages.

Lightweight projection-only adaptation performs competitively with fully supervised systems.

Few-shot adaptation benefits are language-dependent and not strictly increasing with more data.

Abstract

Abusive speech detection is becoming increasingly important as social media shifts towards voice-based interaction, particularly in multilingual and low-resource settings. Most current systems rely on automatic speech recognition (ASR) followed by text-based hate speech classification, but this pipeline is vulnerable to transcription errors and discards prosodic information carried in speech. We investigate whether Contrastive Language-Audio Pre-training (CLAP) can support abusive speech detection directly from audio. Using the ADIMA dataset, we evaluate CLAP-based representations under few-shot supervised contrastive adaptation in cross-lingual and leave-one-language-out settings, with zero-shot prompting included as an auxiliary analysis. Our results show that CLAP yields strong cross-lingual audio representations across ten Indic languages, and that lightweight projection-only adaptation achieves competitive performance with respect to fully supervised systems trained on complete training data. However, the benefits of few-shot adaptation are language-dependent and not monotonic with shot size. These findings suggest that contrastive audio-text models provide a promising basis for cross-lingual audio abuse detection in low-resource settings, while also indicating that transfer remains incomplete and language-specific in important ways.