Commit
•
5c58230
1
Parent(s):
53587f7
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
datasets:
|
3 |
+
- agkphysics/AudioSet
|
4 |
+
- openslr/librispeech_asr
|
5 |
+
pipeline_tag: audio-classification
|
6 |
+
license: bsd-3-clause
|
7 |
+
tags:
|
8 |
+
- audio-classification
|
9 |
+
---
|
10 |
+
|
11 |
+
# Self Supervised Audio Spectrogram Transformer (pretrained on AudioSet/Librispeech)
|
12 |
+
|
13 |
+
Self Supervised Audio Spectrogram Transformer (SSAST) model with uninitialized classifier head. It was introduced in the paper [SSAST: Self-Supervised Audio Spectrogram Transformer](https://arxiv.org/pdf/2110.09784) by Gong et al. and first released in [this repository](https://github.com/YuanGongND/ssast).
|
14 |
+
|
15 |
+
Disclaimer: The team releasing Audio Spectrogram Transformer did not write a model card for this model.
|
16 |
+
|
17 |
+
## Model description
|
18 |
+
|
19 |
+
The Audio Spectrogram Transformer is equivalent to [ViT](https://huggingface.co/docs/transformers/model_doc/vit), but applied on audio. Audio is first turned into an image (as a spectrogram), after which a Vision Transformer is applied. The model gets state-of-the-art results on several audio classification benchmarks.
|
20 |
+
|
21 |
+
## Usage
|
22 |
+
|
23 |
+
The model is pretrained on a massive amount of audio. Please finetune the classifier head before use, as it comes uninitialized.
|
24 |
+
---
|