Update app.py

app.py

@@ -2,9 +2,11 @@ import streamlit as st
 from huggingface_hub import InferenceClient
 from gtts import gTTS
 import base64
+from pydub import AudioSegment
+from pydub.playback import play
 
 client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
-system_prompt = "Tu nombre será Chaman 3.0 una IA conductual, tu rol es la bioética y el existencialismo estóico, holísticamente."
+system_prompt = "Tu nombre será Chaman 3.0 una IA conductual, hablaras español tu rol es la bioética y el existencialismo estóico, holísticamente."
 system_prompt_sent = False
 
 def format_prompt(message, history):
@@ -27,6 +29,12 @@ def text_to_speech(text, speed=2.0):
     tts = gTTS(text=text, lang='es')
     audio_file_path = 'output.mp3'
     tts.save(audio_file_path)
+
+    # Ajustar la velocidad del audio con pydub
+    sound = AudioSegment.from_mp3(audio_file_path)
+    sound = sound.speedup(playback_speed=speed)
+    sound.export(audio_file_path, format="mp3")
+
     return audio_file_path
 
 def generate(user_input, history, temperature=None, max_new_tokens=2048, top_p=0.95, repetition_penalty=1.0):
@@ -54,7 +62,7 @@ def generate(user_input, history, temperature=None, max_new_tokens=2048, top_p=0
     
     response = ' '.join(response.split()).replace('</s>', '')
 
-    audio_file_path = text_to_speech(response)
+    audio_file_path = text_to_speech(response, speed=2.0) 
     audio_file = open(audio_file_path, 'rb')
     audio_bytes = audio_file.read()
 
@@ -73,7 +81,7 @@ st.text_area("Salida del Chatbot", value=output, height=200, max_chars=500, key=
 
 st.markdown(
     f"""
-    <audio autoplay="autoplay" controls="controls" src="data:audio/mp3;base64,{base64.b64encode(audio_bytes).decode()}" type="audio/mp3" speed="2.0" id="audio_player"></audio>
+    <audio autoplay="autoplay" controls="controls" src="data:audio/mp3;base64,{base64.b64encode(audio_bytes).decode()}" type="audio/mp3" id="audio_player"></audio>
     """,
     unsafe_allow_html=True
-)
+)