Synthetic Speech Source Tracing using Metric Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916775178797056 |
|---|---|
| author | Koutsianos, Dimitrios Zacharopoulos, Stavros Panagakis, Yannis Stafylakis, Themos |
| author_facet | Koutsianos, Dimitrios Zacharopoulos, Stavros Panagakis, Yannis Stafylakis, Themos |
| contents | This paper addresses source tracing in synthetic speech-identifying generative systems behind manipulated audio via speaker recognition-inspired pipelines. While prior work focuses on spoofing detection, source tracing lacks robust solutions. We evaluate two approaches: classification-based and metric-learning. We tested our methods on the MLAADv5 benchmark using ResNet and self-supervised learning (SSL) backbones. The results show that ResNet achieves competitive performance with the metric learning approach, matching and even exceeding SSL-based systems. Our work demonstrates ResNet's viability for source tracing while underscoring the need to optimize SSL representations for this task. Our work bridges speaker recognition methodologies with audio forensic challenges, offering new directions for combating synthetic media manipulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_02590 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Synthetic Speech Source Tracing using Metric Learning Koutsianos, Dimitrios Zacharopoulos, Stavros Panagakis, Yannis Stafylakis, Themos Sound Computation and Language This paper addresses source tracing in synthetic speech-identifying generative systems behind manipulated audio via speaker recognition-inspired pipelines. While prior work focuses on spoofing detection, source tracing lacks robust solutions. We evaluate two approaches: classification-based and metric-learning. We tested our methods on the MLAADv5 benchmark using ResNet and self-supervised learning (SSL) backbones. The results show that ResNet achieves competitive performance with the metric learning approach, matching and even exceeding SSL-based systems. Our work demonstrates ResNet's viability for source tracing while underscoring the need to optimize SSL representations for this task. Our work bridges speaker recognition methodologies with audio forensic challenges, offering new directions for combating synthetic media manipulation. |
| title | Synthetic Speech Source Tracing using Metric Learning |
| topic | Sound Computation and Language |
| url | https://arxiv.org/abs/2506.02590 |