Update app.py

app.py

@@ -14,6 +14,7 @@ def video_to_descriptions(video, target_language="en"):
     
     # Load the translation pipeline for the target language
     translator = translation_pipeline("translation", model=f"Helsinki-NLP/opus-mt-en-{target_language}")
+    audio = pipeline("text-to-speech", model="suno/bark")
     
     # Open the video
     cap = cv2.VideoCapture(video)
@@ -49,16 +50,20 @@ def video_to_descriptions(video, target_language="en"):
     
     # Translate the summarized text into the target language
     translated_text = translator(concatenated_descriptions[0]["summarized-text"])[0]["translation_text"]
+
+    audio_file = audio(translated_text)[0]["audio"]
     
-    return translated_text
+    return audio_file
 
 # Create a dropdown menu with language options
-language_dropdown = gr.Interface.dropdown(["en", "fr", "de"], label="Langue cible")
+language_dropdown = gr.Dropdown(
+            ["en", "fr", "de", "es"], label="Language", info="The Language of the output"
+        )
 
 iface = gr.Interface(
     fn=video_to_descriptions,
-    inputs=[gr.Video(type="file", label="Importez une vidéo"), language_dropdown],
-    outputs="text",
+    inputs=[gr.Video(label="Import a Video", info="The Video to be described"), language_dropdown],
+    outputs="audio",
     live=False
 )