ALLM4ADD: Unlocking the Capabilities of Audio Large Language Models for Audio Deepfake Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918085780307968 |
|---|---|
| author | Gu, Hao Yi, Jiangyan Wang, Chenglong Tao, Jianhua Lian, Zheng He, Jiayi Ren, Yong Chen, Yujie Wen, Zhengqi |
| author_facet | Gu, Hao Yi, Jiangyan Wang, Chenglong Tao, Jianhua Lian, Zheng He, Jiayi Ren, Yong Chen, Yujie Wen, Zhengqi |
| contents | Audio deepfake detection (ADD) has grown increasingly important due to the rise of high-fidelity audio generative models and their potential for misuse. Given that audio large language models (ALLMs) have made significant progress in various audio processing tasks, a heuristic question arises: \textit{Can ALLMs be leveraged to solve ADD?}. In this paper, we first conduct a comprehensive zero-shot evaluation of ALLMs on ADD, revealing their ineffectiveness. To this end, we propose ALLM4ADD, an ALLM-driven framework for ADD. Specifically, we reformulate ADD task as an audio question answering problem, prompting the model with the question: ``Is this audio fake or real?''. We then perform supervised fine-tuning to enable the ALLM to assess the authenticity of query audio. Extensive experiments are conducted to demonstrate that our ALLM-based method can achieve superior performance in fake audio detection, particularly in data-scarce scenarios. As a pioneering study, we anticipate that this work will inspire the research community to leverage ALLMs to develop more effective ADD systems. Code is available at https://github.com/ucas-hao/qwen_audio_for_add.git |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_11079 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ALLM4ADD: Unlocking the Capabilities of Audio Large Language Models for Audio Deepfake Detection Gu, Hao Yi, Jiangyan Wang, Chenglong Tao, Jianhua Lian, Zheng He, Jiayi Ren, Yong Chen, Yujie Wen, Zhengqi Sound Computation and Language Audio and Speech Processing Audio deepfake detection (ADD) has grown increasingly important due to the rise of high-fidelity audio generative models and their potential for misuse. Given that audio large language models (ALLMs) have made significant progress in various audio processing tasks, a heuristic question arises: \textit{Can ALLMs be leveraged to solve ADD?}. In this paper, we first conduct a comprehensive zero-shot evaluation of ALLMs on ADD, revealing their ineffectiveness. To this end, we propose ALLM4ADD, an ALLM-driven framework for ADD. Specifically, we reformulate ADD task as an audio question answering problem, prompting the model with the question: ``Is this audio fake or real?''. We then perform supervised fine-tuning to enable the ALLM to assess the authenticity of query audio. Extensive experiments are conducted to demonstrate that our ALLM-based method can achieve superior performance in fake audio detection, particularly in data-scarce scenarios. As a pioneering study, we anticipate that this work will inspire the research community to leverage ALLMs to develop more effective ADD systems. Code is available at https://github.com/ucas-hao/qwen_audio_for_add.git |
| title | ALLM4ADD: Unlocking the Capabilities of Audio Large Language Models for Audio Deepfake Detection |
| topic | Sound Computation and Language Audio and Speech Processing |
| url | https://arxiv.org/abs/2505.11079 |