Upload README.md with huggingface_hub

README.md

@@ -38,44 +38,40 @@ Inspired by [DAC](https://arxiv.org/abs/2306.06546) (Descript Audio Codec). Stri
 
 ![mel spectrogram comparison](https://raw.githubusercontent.com/taresh18/nano-codec/main/assets/aud_8_mel.png)
 
-## 🏃 Quick Start
+## Usage
 
-**1. Clone & Install**
-```bash
-git clone https://github.com/taresh18/nano-codec.git
-cd nano-codec
-uv sync
-```
+```python
+from huggingface_hub import hf_hub_download
+import torch, yaml, soundfile as sf, torchaudio
+from model import RVQCodec
 
-**2. Reconstruct Audio**
-```bash
-cd nano_codec
-python inference.py --input audio.wav --output reconstructed.wav
-```
-Downloads model weights from HuggingFace on first run. Resamples to 16kHz if needed.
+# load model
+model_path = hf_hub_download("taresh18/nano-codec", "model.pt")
+config_path = hf_hub_download("taresh18/nano-codec", "config.yaml")
 
-**3. Train Your Own**
-```bash
-cd nano_codec
-python prepare_data.py    # download LibriSpeech, chunk into shards
-python train.py           # config in configs/config.yaml
-```
+with open(config_path) as f:
+    cfg = yaml.safe_load(f)
+
+model = RVQCodec(in_ch=1, latent_ch=cfg['latent_dim'], K=cfg['codebook_size'],
+                 num_rvq_levels=cfg['num_rvq_levels'], codebook_dim=cfg.get('codebook_dim', 8))
+model.load_state_dict(torch.load(model_path, map_location="cpu", weights_only=True))
+model.eval()
+
+# reconstruct audio
+audio, sr = sf.read("input.wav", dtype="float32")
+waveform = torch.from_numpy(audio).unsqueeze(0).unsqueeze(0)  # [1, 1, T]
+if sr != 16000:
+    waveform = torchaudio.functional.resample(waveform, sr, 16000)
 
-## 🏗️ Project Structure
+with torch.no_grad():
+    recon, _, _, _ = model(waveform)
 
+sf.write("reconstructed.wav", recon[0, 0].numpy(), 16000)
 ```
-nano-codec/
-├── configs/
-│   └── config.yaml           # Training & model config
-├── nano_codec/
-│   ├── model.py              # RVQCodec, VQ, RVQ, encoder/decoder
-│   ├── loss.py               # Multi-scale spectral losses
-│   ├── loader.py             # Dataset loading (in-memory + streaming)
-│   ├── train.py              # Training loop
-│   ├── inference.py          # Reconstruct audio from trained model
-│   ├── prepare_data.py       # Preprocess LibriSpeech into chunks
-│   └── utils.py              # Checkpointing, logging, profiling
-└── assets/                   # Audio samples, images
+
+Or use the inference script from the [GitHub repo](https://github.com/taresh18/nano-codec):
+```bash
+python inference.py --input audio.wav --output reconstructed.wav
 ```
 
 ## 📚 References