This project automates the process of converting email newsletters into audio summaries. It fetches emails from a specified Gmail account, cleans and translates the content, converts it to a high-quality audio file, and sends it to a Telegram chat. This is perfect for listening to your favorite newsletters on the go.

• Email Fetching: Connects to a Gmail account to fetch the latest emails from specified senders.
• Content Processing: Uses OpenAI's GPT models to clean up HTML content and extract the core message.
• Translation: Translates content from one language to another (e.g., English to Italian).
• Text-to-Audio: Converts the processed text into natural-sounding speech using OpenAI's TTS model.
• Audio Speed Control: Adjusts the speed of the generated audio for a customized listening experience (e.g., 1.2x speed).
• Telegram Integration: Automatically sends the final audio file to a specified Telegram chat.
• Caching: Keeps track of processed emails to avoid duplicates.
• Configurable: Easily configure email senders, languages, and API keys.

1. Fetch Email: The script connects to Gmail and retrieves the latest email from a predefined list of senders.
2. Save Raw Content: The raw email text is saved to the data/input_emails directory.
3. Clean & Translate: The text is cleaned to remove boilerplate and then translated using an LLM. Intermediate files are stored in data/cleaned_emails and data/translated_emails.
4. Generate Audio: The translated text is converted to an MP3 audio file.
5. Adjust Speed: The audio speed is increased using pytsmod. The final file is saved in data/generated_audio.
6. Send to Telegram: The MP3 file is sent as an audio message to a Telegram chat.
7. Cache: The email's ID is cached to prevent reprocessing.

1. Clone the repository:

   git clone <repository_url>
   cd <repository_name>

2. Install dependencies: It's recommended to use a virtual environment.

   python -m venv venv
   source venv/bin/activate  # On Windows, use `venv\Scripts\activate`
   pip install -r requirements.txt

3. Create .env file: Create a .env file in the project root and add the following credentials.

   # OpenAI API Key
   OPENAI_API_KEY="your_openai_api_key"
   
   # Gmail Credentials (use an App Password for security)
   EMAIL_ADDRESS="your_email@gmail.com"
   EMAIL_PASSWORD="your_gmail_app_password"
   
   # Telegram Bot Credentials
   TELEGRAM_BOT_TOKEN="your_telegram_bot_token"
   TELEGRAM_CHAT_ID="your_telegram_chat_id"

   Note on Gmail App Password: It is highly recommended to use an App Password for EMAIL_PASSWORD instead of your regular Google account password.

• Allowed Senders: To specify which email newsletters to process, edit the ALLOWED_EMAILS_ADDRESS list in scripts/constants.py.

  # scripts/constants.py
  ALLOWED_EMAILS_ADDRESS = [
      ("dan@tldrnewsletter.com", "TLDR AI"),
  ]

• Audio Speed: You can change the default audio speed by modifying the speed_factor argument in the text_to_audio call within main.py or directly in the text_to_audio.py function definition.

To run the full pipeline, execute the main.py script:

python main.py

The script will log its progress to the console and to log files in the .logs directory.