podcastcreatorapp.py

# -*- coding: utf-8 -*-
"""podcastcreatorapp.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1VN0_bQSwY0tC9PzGMbeU-j7iN4TLbz8C
"""

# Commented out IPython magic to ensure Python compatibility.
!git clone https://github.com/myshell-ai/MeloTTS.git

# %cd MeloTTS

!pip install flask groq requests

!pip install cn2an pypinyin num2words pykakasi mecab-python3 unidic-lite fugashi g2p_en anyascii jamo gruut cached_path

!pip install gradio

# from flask_ngrok import run_with_ngrok
from flask import Flask, request, render_template
import re
import os
import requests
from groq import Groq
from melo.api import TTS

app = Flask(__name__)

# Initialize MeloTTS model
import torch
from melo.api import TTS
from melo.app import speaker_ids
device = 'auto'

model = TTS(language='EN', device=device)
speaker_ids = model.hps.data.spk2id

# Directory to save audio files
AUDIO_DIR = 'static'
os.makedirs(AUDIO_DIR, exist_ok=True)

TEMPLATE_DIR = 'templates'
os.makedirs(TEMPLATE_DIR, exist_ok=True)

# Retrieve the API key from the environment variable
GROQ_API_KEY = os.getenv('GROQ_API_KEY')

# Initialize the Groq client with the API key
groq_client = Groq(api_key=GROQ_API_KEY)

def clean_text_for_tts(text):
    # Replace or remove unwanted characters
    clean_text = re.sub(r'\*\*', '', text)  # Remove double asterisks
    clean_text = re.sub(r'\*', '', clean_text)  # Remove single asterisks
    clean_text = re.sub(r'•', '', clean_text)  # Remove bullet points
    clean_text = re.sub(r'(\d+\.)', '', clean_text)  # Remove numbers followed by a period (like "1.")
    clean_text = re.sub(r'\n', ' ', clean_text)  # Replace newlines with spaces

    # Additional replacements can be done here as needed
    return clean_text

def generate_podcast_content(prompt):
    try:
        response = groq_client.chat.completions.create(
            messages=[
                {"role": "user", "content": prompt}
            ],
            model="llama3-8b-8192"
        )
        content = response.choices[0].message.content
        return content
    except Exception as e:
        print(f"Error in generate_podcast_content: {e}")
        return "Error generating content."

def text_to_speech(text):
    try:
        # Clean the text
        cleaned_text = clean_text_for_tts(text)

        # Initialize MeloTTS model
        model = TTS(language='EN', device=device)

        # Define the path to save the audio file in the static directory
        static_audio_path = os.path.join('static', 'podcast.wav')

        # Convert text to speech
        model.tts_to_file(cleaned_text, static_audio_path, speed=1.0)

        # Return the audio file path
        return static_audio_path
    except Exception as e:
        print(f"Error in text_to_speech: {e}")
        return None

@app.route('/', methods=['GET', 'POST'])
def index():
    if request.method == 'POST':
        prompt_input = request.form['prompt_input']
        content = generate_podcast_content(prompt_input)
        audio_file_path = text_to_speech(content)
        return render_template('result.html', content=content, audio_file_path=audio_file_path)
    return render_template('index.html')

if __name__ == '__main__':
    app.run(debug=True)