Instructions to use mlx-community/dac-speech-24khz-3kbps with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/dac-speech-24khz-3kbps with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir dac-speech-24khz-3kbps mlx-community/dac-speech-24khz-3kbps
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Upload 2 files
Browse files- config.json +16 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"codebook_dim": 8,
|
| 3 |
+
"codebook_loss_weight": 1.0,
|
| 4 |
+
"codebook_size": 1024,
|
| 5 |
+
"commitment_loss_weight": 0.25,
|
| 6 |
+
"decoder_hidden_size": 1536,
|
| 7 |
+
"downsampling_ratios": [2, 4, 5, 8],
|
| 8 |
+
"encoder_hidden_size": 64,
|
| 9 |
+
"hidden_size": 1024,
|
| 10 |
+
"hop_length": 512,
|
| 11 |
+
"model_type": "dac",
|
| 12 |
+
"n_codebooks": 4,
|
| 13 |
+
"quantizer_dropout": 0.0,
|
| 14 |
+
"sampling_rate": 24000,
|
| 15 |
+
"upsampling_ratios": [8, 5, 4, 2]
|
| 16 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43437ddade7c8a57897f1ff4d7adc61d9fa4b4b5f3f324a7fe4c96dd777f8665
|
| 3 |
+
size 295884595
|