Effective Pre-Training of Audio Transformers for Sound Event Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916497811570688 |
|---|---|
| author | Schmid, Florian Morocutti, Tobias Foscarin, Francesco Schlüter, Jan Primus, Paul Widmer, Gerhard |
| author_facet | Schmid, Florian Morocutti, Tobias Foscarin, Francesco Schlüter, Jan Primus, Paul Widmer, Gerhard |
| contents | We propose a pre-training pipeline for audio spectrogram transformers for frame-level sound event detection tasks. On top of common pre-training steps, we add a meticulously designed training routine on AudioSet frame-level annotations. This includes a balanced sampler, aggressive data augmentation, and ensemble knowledge distillation. For five transformers, we obtain a substantial performance improvement over previously available checkpoints both on AudioSet frame-level predictions and on frame-level sound event detection downstream tasks, confirming our pipeline's effectiveness. We publish the resulting checkpoints that researchers can directly fine-tune to build high-performance models for sound event detection tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_09546 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Effective Pre-Training of Audio Transformers for Sound Event Detection Schmid, Florian Morocutti, Tobias Foscarin, Francesco Schlüter, Jan Primus, Paul Widmer, Gerhard Audio and Speech Processing Sound We propose a pre-training pipeline for audio spectrogram transformers for frame-level sound event detection tasks. On top of common pre-training steps, we add a meticulously designed training routine on AudioSet frame-level annotations. This includes a balanced sampler, aggressive data augmentation, and ensemble knowledge distillation. For five transformers, we obtain a substantial performance improvement over previously available checkpoints both on AudioSet frame-level predictions and on frame-level sound event detection downstream tasks, confirming our pipeline's effectiveness. We publish the resulting checkpoints that researchers can directly fine-tune to build high-performance models for sound event detection tasks. |
| title | Effective Pre-Training of Audio Transformers for Sound Event Detection |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2409.09546 |