app.py

import gradio as gr
import requests
import openai
import asyncio
import os
from deepgram import Deepgram
from vocode.streaming.models.transcriber import (
    DeepgramTranscriberConfig,
    PunctuationEndpointingConfig,
)
from vocode.streaming.models.agent import ChatGPTAgentConfig
from vocode.streaming.models.message import BaseMessage
from vocode.streaming.models.synthesizer import ElevenLabsSynthesizerConfig
from vocode.streaming.transcriber.deepgram_transcriber import DeepgramTranscriber
from vocode.streaming.agent.chat_gpt_agent import ChatGPTAgent
from vocode.streaming.synthesizer.eleven_labs_synthesizer import ElevenLabsSynthesizer
from vocode.streaming.streaming_conversation import StreamingConversation
from vocode.helpers import create_streaming_microphone_input_and_speaker_output

# Fetch API keys and voice IDs from environment variables
DEEPGRAM_API_KEY = os.getenv("DEEPGRAM_API_KEY")
ELEVEN_LABS_API_KEY = os.getenv("ELEVEN_LABS_API_KEY")
VOICE_ID = os.getenv("VOICE_ID")
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")

# Initialize OpenAI client
client = openai.OpenAI(api_key=OPENAI_API_KEY)

# Initialize Deepgram
deepgram = Deepgram(DEEPGRAM_API_KEY)

# Function to transcribe audio using Deepgram
async def transcribe_audio(audio_file_path):
    with open(audio_file_path, 'rb') as audio_file:
        audio_data = audio_file.read()

    response = await deepgram.transcription.prerecorded(
        {"buffer": audio_data, "mimetype": "audio/wav"},
        {'punctuate': True, 'language': 'en'}
    )
    transcription = response['results']['channels'][0]['alternatives'][0]['transcript']
    return transcription

# Function to generate content using OpenAI GPT-4
def generate_content(input_text):
    response = client.chat.completions.create(
        model="gpt-4",
        messages=[
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": input_text}
        ]
    )
    generated_text = response.choices[0].message.content.strip()
    return generated_text

# Function to convert text to speech using Eleven Labs
def text_to_speech(text):
    url = f"https://api.elevenlabs.io/v1/text-to-speech/{VOICE_ID}"
    headers = {
        "Accept": "audio/mpeg",
        "Content-Type": "application/json",
        "xi-api-key": ELEVEN_LABS_API_KEY
    }
    data = {
        "text": text,
        "voice_settings": {
            "stability": 0.75,
            "similarity_boost": 0.75
        }
    }
    response = requests.post(url, json=data, headers=headers)
    
    if response.status_code == 200:
        with open("output.mp3", "wb") as f:
            f.write(response.content)
        return "output.mp3"
    else:
        return f"Error: {response.status_code} - {response.text}"

# Main function to handle the entire process
async def process_audio(audio):
    transcription = await transcribe_audio(audio)
    generated_text = generate_content(transcription)
    audio_file = text_to_speech(generated_text)
    return transcription, generated_text, audio_file

# Gradio interface setup
interface = gr.Interface(
    fn=lambda audio: asyncio.run(process_audio(audio)),
    inputs=gr.Audio(type="filepath", label="Speak into your microphone"),
    outputs=[
        gr.Textbox(label="Transcription Output"),
        gr.Textbox(label="Generated Content"),
        gr.Audio(label="Synthesized Speech")
    ],
    title="Speech-to-Text, Content Generation, and Text-to-Speech",
    description="Speak into the microphone, and the system will transcribe your speech, generate content, and convert the generated text into speech."
)

# Launch the Gradio interface
interface.launch()