Fast Text-to-Audio Generation with Adversarial Post-Training
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915293065904128 |
|---|---|
| author | Novack, Zachary Evans, Zach Zukowski, Zack Taylor, Josiah Carr, CJ Parker, Julian Al-Sinan, Adnan Iodice, Gian Marco McAuley, Julian Berg-Kirkpatrick, Taylor Pons, Jordi |
| author_facet | Novack, Zachary Evans, Zach Zukowski, Zack Taylor, Josiah Carr, CJ Parker, Julian Al-Sinan, Adnan Iodice, Gian Marco McAuley, Julian Berg-Kirkpatrick, Taylor Pons, Jordi |
| contents | Text-to-audio systems, while increasingly performant, are slow at inference time, thus making their latency unpractical for many creative applications. We present Adversarial Relativistic-Contrastive (ARC) post-training, the first adversarial acceleration algorithm for diffusion/flow models not based on distillation. While past adversarial post-training methods have struggled to compare against their expensive distillation counterparts, ARC post-training is a simple procedure that (1) extends a recent relativistic adversarial formulation to diffusion/flow post-training and (2) combines it with a novel contrastive discriminator objective to encourage better prompt adherence. We pair ARC post-training with a number optimizations to Stable Audio Open and build a model capable of generating $\approx$12s of 44.1kHz stereo audio in $\approx$75ms on an H100, and $\approx$7s on a mobile edge-device, the fastest text-to-audio model to our knowledge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_08175 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Fast Text-to-Audio Generation with Adversarial Post-Training Novack, Zachary Evans, Zach Zukowski, Zack Taylor, Josiah Carr, CJ Parker, Julian Al-Sinan, Adnan Iodice, Gian Marco McAuley, Julian Berg-Kirkpatrick, Taylor Pons, Jordi Sound Artificial Intelligence Machine Learning Multimedia Audio and Speech Processing Text-to-audio systems, while increasingly performant, are slow at inference time, thus making their latency unpractical for many creative applications. We present Adversarial Relativistic-Contrastive (ARC) post-training, the first adversarial acceleration algorithm for diffusion/flow models not based on distillation. While past adversarial post-training methods have struggled to compare against their expensive distillation counterparts, ARC post-training is a simple procedure that (1) extends a recent relativistic adversarial formulation to diffusion/flow post-training and (2) combines it with a novel contrastive discriminator objective to encourage better prompt adherence. We pair ARC post-training with a number optimizations to Stable Audio Open and build a model capable of generating $\approx$12s of 44.1kHz stereo audio in $\approx$75ms on an H100, and $\approx$7s on a mobile edge-device, the fastest text-to-audio model to our knowledge. |
| title | Fast Text-to-Audio Generation with Adversarial Post-Training |
| topic | Sound Artificial Intelligence Machine Learning Multimedia Audio and Speech Processing |
| url | https://arxiv.org/abs/2505.08175 |