Update app.py

app.py

@@ -3,18 +3,18 @@ import base64
 import io
 from huggingface_hub import InferenceClient
 from gtts import gTTS
-from audiorecorder import audiorecorder
 import speech_recognition as sr
 from pydub import AudioSegment
+from audiorecorder import audiorecorder
 
 if "history" not in st.session_state:
     st.session_state.history = []
 
 def recognize_speech(audio_data, show_messages=True):
     recognizer = sr.Recognizer()
-    
+
     try:
-        audio_data.seek(0)  # Asegurarse de que el puntero esté al inicio del archivo
+        audio_data.seek(0)
         audio = gTTS(audio_data.read(), lang='es')
         audio_recording = sr.AudioFile(io.BytesIO(audio_data.read()))
 
@@ -68,7 +68,7 @@ def generate(audio_text, history, temperature=None, max_new_tokens=256, top_p=0.
 
     for response_token in stream:
         response += response_token.token.text
-    
+
     response = ' '.join(response.split()).replace('</s>', '')
     audio_file = text_to_speech(response)
     return response, audio_file
@@ -99,11 +99,10 @@ def main():
         audio_text = recognize_speech(audio_data)
 
         if audio_text:
-            output, audio_file = generate(audio_text, history=st.session_state.history)  
+            output, audio_file = generate(audio_text, history=st.session_state.history)
 
-            if audio_text:  
-                st.session_state.history.append((audio_text, output))  
-            # respuesta de voz
+            if audio_text:
+                st.session_state.history.append((audio_text, output))
             if audio_file is not None:
                 play_audio(audio_file)