Instructions to use mlx-community/audiogen-medium-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/audiogen-medium-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir audiogen-medium-mlx mlx-community/audiogen-medium-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Fix encodec num_filters: 32 → 64 to match actual weight dimensions (decoder first conv is 1024 channels = 64 * 2^4, not 512)
Browse files- config.json +1 -1
config.json
CHANGED
|
@@ -43,7 +43,7 @@
|
|
| 43 |
"encodec": {
|
| 44 |
"model_type": "encodec",
|
| 45 |
"audio_channels": 1,
|
| 46 |
-
"num_filters":
|
| 47 |
"kernel_size": 7,
|
| 48 |
"num_residual_layers": 1,
|
| 49 |
"dilation_growth_rate": 2,
|
|
|
|
| 43 |
"encodec": {
|
| 44 |
"model_type": "encodec",
|
| 45 |
"audio_channels": 1,
|
| 46 |
+
"num_filters": 64,
|
| 47 |
"kernel_size": 7,
|
| 48 |
"num_residual_layers": 1,
|
| 49 |
"dilation_growth_rate": 2,
|