Update app.py

app.py

@@ -12,8 +12,11 @@ tts = pipeline("text-to-speech", model="Baghdad99/english_voice_tts")
 
 # Define the function to translate speech
 def translate_speech(audio):
+    # Extract the audio data from the tuple
+    audio_data = audio[0]
+
     # Use the speech recognition pipeline to transcribe the audio
-    transcription = pipe(audio)[0]["transcription"]
+    transcription = pipe(audio_data)["transcription"]
 
     # Use the translation pipeline to translate the transcription
     translated_text = translator(transcription, return_tensors="pt", padding=True)
@@ -27,6 +30,7 @@ def translate_speech(audio):
 
     return 16000, synthesised_speech
 
+
 # Define the Gradio interface
 iface = gr.Interface(
     fn=translate_speech,