Audio-to-Audio
MLX
Safetensors
speech-enhancement
speech-denoising
gtcrn
apple-silicon
onnx-conversion
Instructions to use agentable/gtcrn-simple-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use agentable/gtcrn-simple-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir gtcrn-simple-mlx agentable/gtcrn-simple-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
File size: 1,111 Bytes
95fa4d4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 | {
"protocol": "gtcrn-simple-v1",
"model_type": "gtcrn",
"sample_rate": 16000,
"resample": {
"kind": "sherpa-linear-resample",
"lowpass_cutoff_ratio": 0.99,
"lowpass_filter_width": 6,
"flush": true
},
"stft": {
"n_fft": 512,
"hop_length": 256,
"window_length": 512,
"window_type": "hann_sqrt",
"bins": 257
},
"input": {
"name": "mix",
"dtype": "float32",
"shape": [1, 257, 1, 2],
"complex_layout": "real_imag"
},
"caches": [
{
"input_name": "conv_cache",
"output_name": "conv_cache_out",
"dtype": "float32",
"shape": [2, 1, 16, 16, 33]
},
{
"input_name": "tra_cache",
"output_name": "tra_cache_out",
"dtype": "float32",
"shape": [2, 3, 1, 1, 16]
},
{
"input_name": "inter_cache",
"output_name": "inter_cache_out",
"dtype": "float32",
"shape": [2, 1, 33, 16]
}
],
"output": {
"name": "enh",
"dtype": "float32",
"shape": [1, 257, 1, 2],
"complex_layout": "real_imag"
},
"limits": {
"max_output_samples": 4800000
}
}
|