Spaces:
Sleeping
Sleeping
| title: Streaming Vocos Demo | |
| emoji: 🎙️ | |
| colorFrom: blue | |
| colorTo: purple | |
| sdk: gradio | |
| sdk_version: "4.44.1" | |
| python_version: "3.10" | |
| app_file: app.py | |
| pinned: false | |
| # Streaming Vocos Demo (Gradio) | |
| Upload or record an audio clip, then click **Reconstruct** to run it through the Streaming Vocos vocoder. | |
| **UI layout** | |
| - **Left:** input waveform + input mel-spectrogram | |
| - **Right:** reconstructed waveform + reconstructed mel-spectrogram | |
| The Space loads the Lightning checkpoint from the Hugging Face model repo and runs: | |
| - `mel = model.feature_extractor(audio)` | |
| - `audio_out = model(mel)` (optionally in streaming chunk mode) | |
| Model: `warisqr007/StreamingVocos` | |