Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.05139 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909483214569472 |
|---|---|
| author | Tjandra, Andros Wu, Yi-Chiao Guo, Baishan Hoffman, John Ellis, Brian Vyas, Apoorv Shi, Bowen Chen, Sanyuan Le, Matt Zacharov, Nick Wood, Carleigh Lee, Ann Hsu, Wei-Ning |
| author_facet | Tjandra, Andros Wu, Yi-Chiao Guo, Baishan Hoffman, John Ellis, Brian Vyas, Apoorv Shi, Bowen Chen, Sanyuan Le, Matt Zacharov, Nick Wood, Carleigh Lee, Ann Hsu, Wei-Ning |
| contents | The quantification of audio aesthetics remains a complex challenge in audio processing, primarily due to its subjective nature, which is influenced by human perception and cultural context. Traditional methods often depend on human listeners for evaluation, leading to inconsistencies and high resource demands. This paper addresses the growing need for automated systems capable of predicting audio aesthetics without human intervention. Such systems are crucial for applications like data filtering, pseudo-labeling large datasets, and evaluating generative audio models, especially as these models become more sophisticated. In this work, we introduce a novel approach to audio aesthetic evaluation by proposing new annotation guidelines that decompose human listening perspectives into four distinct axes. We develop and train no-reference, per-item prediction models that offer a more nuanced assessment of audio quality. Our models are evaluated against human mean opinion scores (MOS) and existing methods, demonstrating comparable or superior performance. This research not only advances the field of audio aesthetics but also provides open-source models and datasets to facilitate future work and benchmarking. We release our code and pre-trained model at: https://github.com/facebookresearch/audiobox-aesthetics |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_05139 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Meta Audiobox Aesthetics: Unified Automatic Quality Assessment for Speech, Music, and Sound Tjandra, Andros Wu, Yi-Chiao Guo, Baishan Hoffman, John Ellis, Brian Vyas, Apoorv Shi, Bowen Chen, Sanyuan Le, Matt Zacharov, Nick Wood, Carleigh Lee, Ann Hsu, Wei-Ning Sound Machine Learning Audio and Speech Processing The quantification of audio aesthetics remains a complex challenge in audio processing, primarily due to its subjective nature, which is influenced by human perception and cultural context. Traditional methods often depend on human listeners for evaluation, leading to inconsistencies and high resource demands. This paper addresses the growing need for automated systems capable of predicting audio aesthetics without human intervention. Such systems are crucial for applications like data filtering, pseudo-labeling large datasets, and evaluating generative audio models, especially as these models become more sophisticated. In this work, we introduce a novel approach to audio aesthetic evaluation by proposing new annotation guidelines that decompose human listening perspectives into four distinct axes. We develop and train no-reference, per-item prediction models that offer a more nuanced assessment of audio quality. Our models are evaluated against human mean opinion scores (MOS) and existing methods, demonstrating comparable or superior performance. This research not only advances the field of audio aesthetics but also provides open-source models and datasets to facilitate future work and benchmarking. We release our code and pre-trained model at: https://github.com/facebookresearch/audiobox-aesthetics |
| title | Meta Audiobox Aesthetics: Unified Automatic Quality Assessment for Speech, Music, and Sound |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2502.05139 |