Update app.py
Browse files
app.py
CHANGED
|
@@ -18,7 +18,12 @@ model_card = ModelCard(
|
|
| 18 |
description=(
|
| 19 |
"Separates a full music track into its core components using the Demucs model. \n"
|
| 20 |
"Input: A stereo music mix (WAV or MP3) containing multiple instruments or vocals.\n"
|
| 21 |
-
"Output: Four individual audio stems:- Drums, Bass, Vocals, Other-instrumental, each isolated with high fidelity for remixing or analysis."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
),
|
| 23 |
author="Alexandre Défossez, et al.",
|
| 24 |
tags=["demucs", "source-separation", "pyharp", "stems", "multi-output"]
|
|
|
|
| 18 |
description=(
|
| 19 |
"Separates a full music track into its core components using the Demucs model. \n"
|
| 20 |
"Input: A stereo music mix (WAV or MP3) containing multiple instruments or vocals.\n"
|
| 21 |
+
"Output: Four individual audio stems:- Drums, Bass, Vocals, Other-instrumental, each isolated with high fidelity for remixing or analysis. \n\n"
|
| 22 |
+
"**Model Variants and Recommendations:**\n"
|
| 23 |
+
"1. **htdemucs**: Best overall realism and clarity (default)\n"
|
| 24 |
+
"2. **mdx_extra_q**: Highest vocal quality and cleaner separation\n"
|
| 25 |
+
"3. **mdx_extra**: Balanced trade-off between speed and accuracy\n"
|
| 26 |
+
"4. **mdx_q**: Lightweight and fast, suitable for quick previews or limited hardware\n\n"
|
| 27 |
),
|
| 28 |
author="Alexandre Défossez, et al.",
|
| 29 |
tags=["demucs", "source-separation", "pyharp", "stems", "multi-output"]
|