# Beyond Content: A Comprehensive Speech Toxicity Dataset and Detection Framework Incorporating Paralinguistic Cues

arXiv: https://arxiv.org/abs/2605.15984 · HTML: https://mlsanity.com/paper/2605.15984/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 33 references that are in the map.
- First posted on arXiv 15 May 2026.

## Builds on (most cited first, 18 of 18)

- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations](https://mlsanity.com/paper/1810.02508/) (2018)
- [F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching](https://mlsanity.com/paper/2410.06885/) (2024)
- [ShieldGemma: Generative AI Content Moderation Based on Gemma](https://mlsanity.com/paper/2407.21772/) (2024)
- [SSAST: Self-Supervised Audio Spectrogram Transformer](https://mlsanity.com/paper/2110.09784/) (2021)
- [E2 TTS: Embarrassingly Easy Fully Non-Autoregressive Zero-Shot TTS](https://mlsanity.com/paper/2406.18009/) (2024)
- [Reinforcement Learning Outperforms Supervised Fine-Tuning: A Case Study on Audio Question Answering](https://mlsanity.com/paper/2503.11197/) (2025)
- [Handling Bias in Toxic Speech Detection: A Survey](https://mlsanity.com/paper/2202.00126/) (2022)
- [Emotion Based Hate Speech Detection using Multimodal Learning](https://mlsanity.com/paper/2202.06218/) (2022)
- [DeToxy: A Large-Scale Multimodal Dataset for Toxicity Classification in Spoken Utterances](https://mlsanity.com/paper/2110.07592/) (2021)
- [Enhancing Automated Audio Captioning via Large Language Models with Optimized Audio Encoding](https://mlsanity.com/paper/2406.13275/) (2024)
- [Lightweight Toxicity Detection in Spoken Language: A Transformer-based Approach for Edge Devices](https://mlsanity.com/paper/2304.11408/) (2023)
- [Enhancing Multilingual Voice Toxicity Detection with Speech-Text Alignment](https://mlsanity.com/paper/2406.10325/) (2024)

## Abstract (arXiv)

Toxic speech detection has become a crucial challenge in maintaining safe online communication environments. However, existing approaches to toxic speech detection often neglect the contribution of paralinguistic cues, such as emotion, intonation, and speech rate, which are key to detecting speech toxicity. Moreover, current toxic speech datasets are predominantly text-based, limiting the development of models that can capture paralinguistic cues.To address these challenges, we present ToxiAlert-Bench, a large-scale audio dataset comprising over 30,000 audio clips annotated with seven major toxic categories and twenty fine-grained toxic labels. Uniquely, our dataset annotates toxicity sources -- distinguishing between textual content and paralinguistic origins -- for comprehensive toxic speech analysis.Furthermore, we propose a dual-head neural network with a multi-stage training strategy tailored for toxic speech detection. This architecture features two task-specific classification headers: one for identifying the source of sensitivity (textual or paralinguistic), and the other for categorizing the specific toxic type. The training process involves independent head training followed by joint fine-tuning to reduce task interference. To mitigate data class imbalance, we incorporate class-balanced sampling and weighted loss functions.Our experimental results show that leveraging paralinguistic features significantly improves detection performance. Our method consistently outperforms existing baselines across multiple evaluation metrics, with a 21.1% relative improvement in Macro-F1 score and a 13.0% relative gain in accuracy over the strongest baseline, highlighting its enhanced effectiveness and practical applicability.