app.py

# -*- coding: utf-8 -*-
"""AD.ipynb
Automatically generated by Colaboratory.
Original file is located at
    https://colab.research.google.com/drive/17QP7dk4lEklxpGBT704p1j_yUo1U9EuJ
"""


import gradio as gr
import openai
from elevenlabslib import *
from pydub import AudioSegment
from pydub.playback import play
import io
import config
import os
import librosa
import IPython.display as ipd
from elevenlabslib import ElevenLabsUser

os.system("sudo apt-get install portaudio19-dev")
os.system("pip install whisper openai elevenlabslib pydub pyaudio gradio config")

openai.api_key = os.environ['openapi']
user = ElevenLabsUser("b73020f58fc303e3c343782b85cf7138")
           
messages = ["You are an advisor. Please respond to all input in 50 words or less."]
           
def transcribe(audio , voice_name):
           
    global messages
           
    audio_file = open(audio, "rb")
    transcript = openai.Audio.transcribe("whisper-1", audio_file)
    messages.append(f"\nUser: {transcript['text']}")
            
    response = openai.ChatCompletion.create(
        model="gpt-4",
        messages=[{"role": "user", "content": f"{messages[-1]}"},]
    )
    
    system_message = response["choices"][0]["message"]["content"]
    
    print(system_message)
    
    messages.append(f"{system_message}")
             
    voice = user.get_voices_by_name(voice_name)[0]
    audio = voice.generate_audio_bytes(system_message)
            
    audio = AudioSegment.from_file(io.BytesIO(audio), format="mp3")
    audio.export("output.wav", format="wav")
           
    chat_transcript = "\n".join(messages)
     
    return "output.wav"
              
iface = gr.Interface(fn = transcribe,
                     inputs = [gr.Audio(source="microphone", type="filepath"), "text"],
                     outputs = [gr.Audio()],
                     )
    
iface.launch(share=True , debug=True)