Update app.py

app.py

@@ -3,9 +3,8 @@ import base64
 import io
 from huggingface_hub import InferenceClient
 from gtts import gTTS
-from audiorecorder import audiorecorder
+import audiorecorder
 import speech_recognition as sr
-from pydub import AudioSegment
 
 pre_prompt_text = "You are a behavioral AI, your answers should be brief, stoic and humanistic."
 
@@ -74,23 +73,18 @@ def generate(audio_text, history, temperature=None, max_new_tokens=512, top_p=0.
         response += response_token.token.text
     
     response = ' '.join(response.split()).replace('</s>', '')
-    audio_file = text_to_speech(response, speed=1.3)
+    audio_file = text_to_speech(response)
     return response, audio_file
 
-def text_to_speech(text, speed=1.3):
-    tts = gTTS(text=text, lang='es-ES-Wavenet-B')
+def text_to_speech(text):
+    tts = gTTS(text=text, lang='es')
     audio_fp = io.BytesIO()
     tts.write_to_fp(audio_fp)
     audio_fp.seek(0)
-    audio = AudioSegment.from_file(audio_fp, format="mp3")
-    modified_speed_audio = audio.speedup(playback_speed=speed)
-    modified_audio_fp = io.BytesIO()
-    modified_speed_audio.export(modified_audio_fp, format="mp3")
-    modified_audio_fp.seek(0)
-    return modified_audio_fp
+    return audio_fp
 
 def main():
-    audio_data = audiorecorder("Push to Talk", "Stop Recording...")
+    audio_data = audiorecorder.audiorecorder("Push to Talk", "Stop Recording...")
 
     if not audio_data.empty():
         st.audio(audio_data.export().read(), format="audio/wav")