Spaces:
Runtime error
Runtime error
include references to huggingface model and space
Browse files- README.md +2 -15
- notebooks/test-model.ipynb +0 -0
README.md
CHANGED
|
@@ -22,6 +22,8 @@ Audio can be represented as images by transforming to a [mel spectrogram](https:
|
|
| 22 |
|
| 23 |
A DDPM model is trained on a set of mel spectrograms that have been generated from a directory of audio files. It is then used to synthesize similar mel spectrograms, which are then converted back into audio. See the `test-model.ipynb` notebook for an example.
|
| 24 |
|
|
|
|
|
|
|
| 25 |
## Generate Mel spectrogram dataset from directory of audio files
|
| 26 |
#### Training can be run with Mel spectrograms of resolution 64x64 on a single commercial grade GPU (e.g. RTX 2080 Ti). The `hop_length` should be set to 1024 for better results.
|
| 27 |
|
|
@@ -92,18 +94,3 @@ accelerate launch --config_file accelerate_sagemaker.yaml \
|
|
| 92 |
--lr_warmup_steps 500 \
|
| 93 |
--mixed_precision no
|
| 94 |
```
|
| 95 |
-
=======
|
| 96 |
-
---
|
| 97 |
-
title: Audio Diffusion
|
| 98 |
-
emoji: 📉
|
| 99 |
-
colorFrom: pink
|
| 100 |
-
colorTo: blue
|
| 101 |
-
sdk: gradio
|
| 102 |
-
sdk_version: 3.1.4
|
| 103 |
-
app_file: app.py
|
| 104 |
-
pinned: false
|
| 105 |
-
license: gpl-3.0
|
| 106 |
-
---
|
| 107 |
-
|
| 108 |
-
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
| 109 |
-
>>>>>>> 76320e6ba1ee26f1d98e3cfb63cc5c057b823319
|
|
|
|
| 22 |
|
| 23 |
A DDPM model is trained on a set of mel spectrograms that have been generated from a directory of audio files. It is then used to synthesize similar mel spectrograms, which are then converted back into audio. See the `test-model.ipynb` notebook for an example.
|
| 24 |
|
| 25 |
+
You can play around with the model I trained on about 500 songs from my Spotify "liked" playlist [here](https://huggingface.co/spaces/teticio/audio-diffusion)
|
| 26 |
+
|
| 27 |
## Generate Mel spectrogram dataset from directory of audio files
|
| 28 |
#### Training can be run with Mel spectrograms of resolution 64x64 on a single commercial grade GPU (e.g. RTX 2080 Ti). The `hop_length` should be set to 1024 for better results.
|
| 29 |
|
|
|
|
| 94 |
--lr_warmup_steps 500 \
|
| 95 |
--mixed_precision no
|
| 96 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
notebooks/test-model.ipynb
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|