Update app.py

app.py

@@ -1,3 +1,4 @@
+# gradio app
 import torch
 from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
 import gradio as gr
@@ -42,13 +43,13 @@ def extract_audio_from_video(video_path, audio_output_path):
 def speech_to_text(input_file):
     try:
         if input_file.name.endswith((".mp4", ".avi", ".mov")):
-            audio_file_path = extract_audio_from_video(input_file.name, "temp_audio.mp3")
+            audio_file_path = extract_audio_from_video(input_file, "temp_audio.mp3")
             if audio_file_path:
                 result = pipe(audio_file_path)
-                return result[0]["transcription"]
+                return result["text"]
         else:
-            result = pipe(input_file.read())
-            return result[0]["transcription"]
+            result = pipe(input_file)
+            return result["text"]
     except Exception as e:
         return f"Error: {str(e)}"