Spooftral: Can Voxtral Audio-Language Model Detect Speech Spoofing?
Not on the map. Why?
Abstract (arXiv)
Self-supervised learning (SSL) countermeasures (CMs) have shown strong performance in recent years. However, they often show degraded performance while facing unseen spoofing attacks and mismatched conditions. This study examines the Voxtral audio-language model (ALM) framework for spoofing detection, as a step toward combining CM capabilities within the ALM framework. We analyze how Voxtral captures spoofing cues through audio-text processing and propose an instruction-guided approach that uses label-sequence likelihoods to evaluate bonafide and spoofed speech. Experiments on the ASVspoof databases show that without task-specific adaptation, the LLM layers emphasize semantic representations, reducing the separability of spoof-discriminative acoustic cues compared to the Whisper-based audio encoder. Consequently, spoofing-related information becomes less separable after language-model processing. We also applied lightweight adaptation using weight-decomposed low-rank adaptation (DoRA) to the Voxtral model and propose the Spooftral model, achieving an equal error rate (EER) of 4.25% on the ASVspoof5 evaluation set.
Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 48 of its 67 references that are in the map. First posted on arXiv 23 Sep 2026.
Builds on
48 of 67 references.
References · 48
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,231
- 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,986
- 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,004
- 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,870
- 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,313
- 2021WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processingcited by 1,625
- 2024DoRA: Weight-Decomposed Low-Rank Adaptationcited by 747
- 2022Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentationcited by 713
- 2023Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Modelscited by 601
- 2023SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilitiescited by 523
- 2023SALMONN: Towards Generic Hearing Abilities for Large Language Modelscited by 498
- 2024LoRA+: Efficient Low Rank Adaptation of Large Modelscited by 369
36 more
- 2021YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyonecited by 278
- 2024PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language Modelscited by 265
- 2024Direct Language Model Alignment from Online AI Feedbackcited by 257
- 2019A Study of BFLOAT16 for Deep Learning Trainingcited by 232
- 2023Pengi: An Audio Language Model for Audio Taskscited by 219
- 2025Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Modelscited by 213
- 2023OneLLM: One Framework to Align All Modalities with Languagecited by 185
- 2020Understanding How Dimension Reduction Tools Work: An Empirical Approach to Deciphering t-SNE, UMAP, TriMAP, and PaCMAP for Data Visualizationcited by 170
- 2022Does Audio Deepfake Detection Generalize?cited by 130
- 2022ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wildcited by 118
- 2022Automatic speaker verification spoofing and deepfake detection using wav2vec 2.0 and data augmentationcited by 112
- 2024ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scalecited by 89
- 2024A Survey on Speech Large Language Models for Understandingcited by 71
- 2025Audio-CoT: Exploring Chain-of-Thought Reasoning in Large Audio Language Modelcited by 60
- 2021StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversioncited by 57
- 2023LLaSM: Large Language and Speech Modelcited by 56
- 2025Voxtralcited by 43
- 2023Speech Self-Supervised Representation Benchmarking: Are We Doing it Right?cited by 28
- 2023Improved DeepFake Detection Using Whisper Featurescited by 28
- 2025ASVspoof 5: Design, Collection and Validation of Resources for Spoofing, Deepfake, and Adversarial Attack Detection Using Crowdsourced Speechcited by 25
- 2023Towards generalisable and calibrated synthetic speech detection with self-supervised representationscited by 23
- 2025ALLM4ADD: Unlocking the Capabilities of Audio Large Language Models for Audio Deepfake Detectioncited by 17
- 2024AdaRankGrad: Adaptive Gradient-Rank and Moments for Memory-Efficient LLMs Training and Fine-Tuningcited by 16
- 2023The Impact of Silence on Speech Anti-Spoofingcited by 13
- 2025UALM: Unified Audio Language Model for Understanding, Generation and Reasoningcited by 10
- 2024WHISMA: A Speech-LLM to Perform Zero-shot Spoken Language Understandingcited by 8
- 2024USTC-KXDIGIT System Description for ASVspoof5 Challengecited by 7
- 2025Overcoming Data Scarcity in Multi-Dialectal Arabic ASR via Whisper Fine-Tuningcited by 7
- 2026Voxtral Realtimecited by 6
- 2024Learn from Real: Reality Defender's Submission to ASVspoof5 Challengecited by 5
- 2025Improving Out-of-Domain Audio Deepfake Detection via Layer Selection and Fusion of SSL-Based Countermeasurescited by 3
- 2024Tandem spoofing-robust automatic speaker verification based on time-domain embeddingscited by 2
- 2025Can Quantized Audio Language Models Perform Zero-Shot Spoofing Detection?cited by 2
- 2025A Survey of Threats Against Voice Authentication and Anti-Spoofing Systemscited by 2
- 2025Competitive Audio-Language Models with Data-Efficient Single-Stage Training on Public Datacited by 2
- 2024ASASVIcomtech: The Vicomtech-UGR Speech Deepfake Detection and SASV Systems for the ASVspoof5 Challengecited by 1
Built on by
Not cited yet.