Update app.py

app.py

@@ -7,7 +7,10 @@ import requests
 import base64
 import math
 
+# Erlaubte Dateiformate
 allowed_medias = [".png", ".jpg", ".jpeg", ".bmp", ".gif", ".tiff"]
+allowed_audios = [".mp3", ".wav", ".m4a", ".ogg"]
+
 API_URL = "https://text.pollinations.ai/openai"
 
 def convert_to_wav(audio_path):
@@ -52,8 +55,8 @@ def generate_slideshow_with_audio(images, audio_file, duration_per_image=3, y_po
         segments_per_image = math.ceil(total_words / len(images))
         texts = []
         for i in range(len(images)):
-            start = i*segments_per_image
-            end = min((i+1)*segments_per_image, total_words)
+            start = i * segments_per_image
+            end = min((i + 1) * segments_per_image, total_words)
             texts.append(" ".join(words[start:end]))
     else:
         texts = [""] * len(images)
@@ -69,7 +72,8 @@ def generate_slideshow_with_audio(images, audio_file, duration_per_image=3, y_po
         )
 
         if text:
-            safe_text = text.replace(":", "\\:").replace("'", "\\'")
+            # Escape problematische Zeichen
+            safe_text = text.replace(":", "\\:").replace("'", "\\'").replace(",", "\\,")
             drawtext_filter = (
                 f",drawtext=text='{safe_text}':fontcolor=white:fontsize={font_size}:borderw=2:"
                 f"x=(w-text_w)/2:y=(h-text_h)*{y_pos}:"
@@ -141,7 +145,10 @@ with gr.Blocks() as demo:
     gr.Markdown("# Slideshow mit Audio & automatischen Untertiteln")
 
     img_input = gr.Files(label="Bilder auswählen (mehrere)", file_types=allowed_medias)
-    audio_input = gr.File(label="Audio hinzufügen (MP3, WAV, M4A, OGG ... optional)")
+    audio_input = gr.File(
+        label="Audio hinzufügen (MP3, WAV, M4A, OGG ... optional)",
+        file_types=allowed_audios
+    )
     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)
     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)")
@@ -156,4 +163,4 @@ with gr.Blocks() as demo:
         outputs=[out_video, status]
     )
 
-demo.launch()
+demo.launch()