Update app.py

app.py

@@ -100,22 +100,83 @@ def generate_slideshow_with_audio(images, input_text, duration_per_word=0.5, dur
     except subprocess.CalledProcessError as e:
         return None, f"❌ FFmpeg Concat Fehler:\n{e.stderr}"
 
-    # Text separat anzeigen
-    # Berechnung der Zeiten für jedes Wort
-    temp_text_file = Path(temp_dir) / "text.txt"
-    with open(temp_text_file, "w") as f:
-        for i, word in enumerate(words):
-            f.write(f"file '{str(output_file)}'\n")
-            f.write(f"duration {duration_per_word}\n")
-            f.write(f"word {word}\n")
+    # Text als Overlay Wort für Wort einfügen
+    total_duration = duration_per_image * len(images)  # Gesamtzeit des Videos
+    word_display_time = duration_per_word  # Zeit pro Wort (in Sekunden)
+    
+    clips_with_text = []
+    word_index = 0
+
+    for i, img_path in enumerate(images):
+        img_path = Path(img_path.name)  # Pfad des Bildes
+        clip_path_with_text = Path(temp_dir) / f"clip_with_text_{i}.mp4"
+        
+        # Berechne Start- und Endzeit für jedes Wort
+        start_time = i * duration_per_image
+        end_time = (i + 1) * duration_per_image
+
+        # Erstelle Text-Filters für jedes Wort
+        if word_index < len(words):
+            text = words[word_index]
+            word_index += 1
+        else:
+            text = ""  # Falls keine weiteren Wörter, leeres Text
+
+        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"
+            f",drawtext=text='{shlex.quote(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_per_image}-{fade_duration}), 1, ({duration_per_image}-t)/{fade_duration}))'"
+        )
+
+        # Bild als Video mit Text Overlay erstellen
+        cmd = [
+            "ffmpeg",
+            "-y",
+            "-loop", "1",
+            "-i", str(img_path),
+            "-t", str(duration_per_image),
+            "-vf", vf_filters,
+            str(clip_path_with_text)
+        ]
+        try:
+            subprocess.run(cmd, check=True, capture_output=True, text=True)
+        except subprocess.CalledProcessError as e:
+            return None, f"❌ FFmpeg Fehler bei Text Overlay für Bild {i+1}:\n{e.stderr}"
+
+        clips_with_text.append(clip_path_with_text)
+
+    # Zusammenfügen der Clips mit Text
+    filelist_with_text_path = Path(temp_dir) / "filelist_with_text.txt"
+    with open(filelist_with_text_path, "w") as f:
+        for clip in clips_with_text:
+            f.write(f"file '{clip}'\n")
+
+    output_with_text_file = Path(temp_dir) / f"slideshow_with_text_{uuid.uuid4().hex}.mp4"
+    cmd_concat_with_text = [
+        "ffmpeg",
+        "-y",
+        "-f", "concat",
+        "-safe", "0",
+        "-i", str(filelist_with_text_path),
+        "-c:v", "libx264",
+        "-pix_fmt", "yuv420p",
+        str(output_with_text_file)
+    ]
+    try:
+        subprocess.run(cmd_concat_with_text, check=True, capture_output=True, text=True)
+    except subprocess.CalledProcessError as e:
+        return None, f"❌ FFmpeg Fehler beim Zusammenfügen der Clips mit Text:\n{e.stderr}"
 
     # Audio hinzufügen, falls vorhanden
     if temp_audio_file:
-        final_output = Path(temp_dir) / f"slideshow_audio_{uuid.uuid4().hex}.mp4"
+        final_output = Path(temp_dir) / f"slideshow_with_audio_{uuid.uuid4().hex}.mp4"
         cmd_audio = [
             "ffmpeg",
             "-y",
-            "-i", str(output_file),
+            "-i", str(output_with_text_file),
             "-i", str(temp_audio_file),
             "-c:v", "copy",
             "-c:a", "aac",
@@ -124,15 +185,15 @@ def generate_slideshow_with_audio(images, input_text, duration_per_word=0.5, dur
         ]
         try:
             subprocess.run(cmd_audio, check=True, capture_output=True, text=True)
-            return str(final_output), "✅ Slideshow mit Audio und Text erstellt"
+            return str(final_output), "✅ Slideshow mit Audio und Text Overlay erstellt"
         except subprocess.CalledProcessError as e:
-            return None, f"❌ FFmpeg Audio Merge Fehler:\n{e.stderr}"
+            return None, f"❌ FFmpeg Fehler beim Hinzufügen des Audios:\n{e.stderr}"
 
-    return str(output_file), "✅ Slideshow erstellt (ohne Audio)"
+    return str(output_with_text_file), "✅ Slideshow mit Text Overlay erstellt (ohne Audio)"
 
 # Gradio UI
 with gr.Blocks() as demo:
-    gr.Markdown("# Slideshow mit Audio und separatem Text")
+    gr.Markdown("# Slideshow mit Audio und Text Overlay")
 
     img_input = gr.Files(label="Bilder auswählen (mehrere)", file_types=allowed_medias)
     text_input = gr.Textbox(