SEED: Speaker Embedding Enhancement Diffusion Model
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915298661105664 |
|---|---|
| author | Nam, KiHyun Heo, Jungwoo Jung, Jee-weon Park, Gangin Jung, Chaeyoung Yu, Ha-Jin Chung, Joon Son |
| author_facet | Nam, KiHyun Heo, Jungwoo Jung, Jee-weon Park, Gangin Jung, Chaeyoung Yu, Ha-Jin Chung, Joon Son |
| contents | A primary challenge when deploying speaker recognition systems in real-world applications is performance degradation caused by environmental mismatch. We propose a diffusion-based method that takes speaker embeddings extracted from a pre-trained speaker recognition model and generates refined embeddings. For training, our approach progressively adds Gaussian noise to both clean and noisy speaker embeddings extracted from clean and noisy speech, respectively, via forward process of a diffusion model, and then reconstructs them to clean embeddings in the reverse process. While inferencing, all embeddings are regenerated via diffusion process. Our method needs neither speaker label nor any modification to the existing speaker recognition pipeline. Experiments on evaluation sets simulating environment mismatch scenarios show that our method can improve recognition accuracy by up to 19.6% over baseline models while retaining performance on conventional scenarios. We publish our code here https://github.com/kaistmm/seed-pytorch |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_16798 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SEED: Speaker Embedding Enhancement Diffusion Model Nam, KiHyun Heo, Jungwoo Jung, Jee-weon Park, Gangin Jung, Chaeyoung Yu, Ha-Jin Chung, Joon Son Audio and Speech Processing Artificial Intelligence A primary challenge when deploying speaker recognition systems in real-world applications is performance degradation caused by environmental mismatch. We propose a diffusion-based method that takes speaker embeddings extracted from a pre-trained speaker recognition model and generates refined embeddings. For training, our approach progressively adds Gaussian noise to both clean and noisy speaker embeddings extracted from clean and noisy speech, respectively, via forward process of a diffusion model, and then reconstructs them to clean embeddings in the reverse process. While inferencing, all embeddings are regenerated via diffusion process. Our method needs neither speaker label nor any modification to the existing speaker recognition pipeline. Experiments on evaluation sets simulating environment mismatch scenarios show that our method can improve recognition accuracy by up to 19.6% over baseline models while retaining performance on conventional scenarios. We publish our code here https://github.com/kaistmm/seed-pytorch |
| title | SEED: Speaker Embedding Enhancement Diffusion Model |
| topic | Audio and Speech Processing Artificial Intelligence |
| url | https://arxiv.org/abs/2505.16798 |