Update app.py

app.py

@@ -10,7 +10,7 @@ def generate_slideshow(images, duration, texts=None, y_pos=0.5, fade_duration=0.
     if not images:
         return None, "❌ Keine Bilder ausgewählt"
 
-    y_pos = min(max(0.0, y_pos), 0.9)
+    y_pos = min(max(0.0, y_pos), 0.9)  # 0=oben, 1=unten, max 0.9
     temp_dir = tempfile.mkdtemp()
     clips = []
 
@@ -21,18 +21,22 @@ def generate_slideshow(images, duration, texts=None, y_pos=0.5, fade_duration=0.
         clip_path = Path(temp_dir) / f"clip_{i}.mp4"
         text = texts[i] if i < len(texts) else ""
 
-        # Scale + Pad Filter
-        vf_filters = "scale='min(1280,iw)':-2,pad=1280:720:(1280-iw)/2:(720-ih)/2:color=black,fps=25,format=yuv420p"
+        # Scale + Pad für proportionalen Fit in 1280x720
+        vf_filters = (
+            "scale=w=1280:h=720:force_original_aspect_ratio=decrease,"
+            "pad=1280:720:(ow-iw)/2:(oh-ih)/2:color=black,"
+            "fps=25,format=yuv420p"
+        )
 
         # Text Overlay Filter
         if text:
             safe_text = text.replace(":", "\\:").replace("'", "\\'")
             drawtext_filter = (
-                f"drawtext=text='{safe_text}':fontcolor=white:fontsize={font_size}:borderw=2:"
+                f",drawtext=text='{safe_text}':fontcolor=white:fontsize={font_size}:borderw=2:"
                 f"x=(w-text_w)/2:y=(h-text_h)*{y_pos}:"
                 f"alpha='if(lt(t,{fade_duration}), t/{fade_duration}, if(lt(t,{duration}-{fade_duration}), 1, ({duration}-t)/{fade_duration}))'"
             )
-            vf_filters += f",{drawtext_filter}"
+            vf_filters += drawtext_filter
 
         cmd = [
             "ffmpeg",
@@ -50,6 +54,7 @@ def generate_slideshow(images, duration, texts=None, y_pos=0.5, fade_duration=0.
 
         clips.append(clip_path)
 
+    # Alle Clips zu einem Video zusammenfügen
     filelist_path = Path(temp_dir) / "filelist.txt"
     with open(filelist_path, "w") as f:
         for clip in clips:
@@ -84,7 +89,7 @@ with gr.Blocks() as demo:
     duration_input = gr.Number(value=3, label="Dauer pro Bild in Sekunden", precision=1)
     fade_input = gr.Number(value=0.7, label="Fade Dauer in Sekunden", precision=1)
     text_input = gr.Textbox(label="Texte pro Bild (mit Komma trennen)", placeholder="Text1, Text2, Text3 ...")
-    ypos_input = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, value=0.5, label="Y-Position für alle Texte (0=oben, 0.5=mitte, 1=unten, max 0.9)")
+    ypos_input = gr.Slider(minimum=0.0, maximum=0.9, step=0.01, value=0.5, label="Y-Position für alle Texte (0=oben, 0.5=mitte, 0.9=unten)")
     font_size_input = gr.Number(value=60, label="Textgröße (px)")
     out_video = gr.Video(interactive=False, label="Generiertes Video")
     status = gr.Textbox(interactive=False, label="Status")