description and hints

app.py

@@ -15,7 +15,11 @@ import numpy as np
 # ModelCard
 model_card = ModelCard(
     name="Demucs Stem Separator (All Stems)",
-    description="Separates a music mixture into all individual stems using a Demucs model.",
+    description=(
+    "Separates a full music track into its core components using the Demucs model. \n"
+    "Input: A stereo music mix (WAV or MP3) containing multiple instruments or vocals.\n"
+    "Output: Four individual audio stems — Drums, Bass, Vocals, and Other-Instrumental, each isolated with high fidelity for remixing or analysis."
+    ),
     author="Alexandre Défossez, et al.",
     tags=["demucs", "source-separation", "pyharp", "stems", "multi-output"]
 )
@@ -112,7 +116,14 @@ with gr.Blocks() as demo:
     model_dropdown = gr.Dropdown(
         choices=DEMUX_MODELS,
         label="Demucs Model",
-        value="htdemucs"
+        value="htdemucs",
+        info=(
+            "Choose which Demucs variant to use for separation:\n"
+            "1. htdemucs: Best overall quality and realism (default).\n"
+            "2. mdx_extra_q: High-quality extended model trained for cleaner vocals.\n"
+            "3. mdx_extra: Balanced version with good speed and quality.\n"
+            "4. mdx_q: Faster, lighter model for quick previews or low-end devices."
+        )
     )
 
     output_drums = gr.Audio(type="filepath", label="Drums")