GastappBot: Telegram Bot for Expense Management

This repository contains a Telegram bot called GastappBot that interacts with an expense management system. The bot allows users to ask questions about their expenses and receives progressive responses, mimicking a more conversational flow.

---

🚀 Features

• Expense Management: You can easily query details about your expenses.
• Progressive Responses: The bot simulates an interactive conversation by showing responses in real-time, with the bot "typing..." during the conversation.
• External API Integration: The bot generates responses from an API configured in a JSON file.

---

🛠️ Requirements

Make sure you have the following installed:

• Python 3.11+
• Required libraries:
  • gspread
  • oauth2client
  • pandas
  • typer
  • python-telegram-bot
  • openai-whisper
  • ffmpeg-python

---

⚙️ Installation

1. Clone the repository:

   git clone https://github.com/yourusername/gastappbot.git
   cd gastappbot

2. Create a virtual environment and install dependencies:

   conda create --name [NAME] python=3.11
   conda activate [NAME]

3. Install the dependencies:

   pip install -r requirements.txt

---

🔑 Configuration

1. API Keys:
   Create api_credentials.json file in the credentials/ folder with the following format:

   {
     "openrouter_key": "your_api_key",
     "telegram": "your_bot_key",
     "spreadsheet_id": "your_spreadsheet_id"
   }

2. Spreadsheet credentials:
   Create credentials.json file in the credentials/ folder (see info).

3. Specify your personal configuration:
   Update the config.json file in the root directory with your specific configuration details. The file should include the following fields:

   • sheet_data_name: The name of the sheet in your spreadsheet where the data is stored.
   • prompt_model: The path to the prompt generation class.
   • voice: The path to the voice-to-text processing class.
   • openrouter_models: A list of models to be used by the bot.
   • question: The template for formatting questions.
   • prompt: The template for generating the prompt used by the bot.

   Example config.json:

   {
       "sheet_data_name": "Datos",
       "prompt_model": "prompt.prompt.MyPrompt",
       "voice": "voice.voicetotext.WhisperVoiceToText",
       "openrouter_models": [
         {"name": "deepseek/deepseek-r1:free"},
         {"name": "deepseek/deepseek-chat:free"}
       ],
       "question": "Pregunta: {question}.",
       "prompt": "..."
   }

   You can create new voice models, NLP models, and prompts in the corresponding folders and use them by modifying the config.json file.

---

📝 Usage

1. Set up the Telegram Bot:
   Create a bot on Telegram using BotFather and get your token (update the api_credentials.json file).

2. Update the data:

   python update_data.py

3. Run the bot:
   To start the bot, run the following command:

   python telegram_bot.py

4. Interact with the bot:
   On Telegram, search for your bot and start chatting with it by asking questions related to your expenses.

---

📚 Project Structure

The repository contains the following folders and files:

/gastappbot
  ├── /constants             # Folder for the constants
      └── constants.py
  ├── /prompt                # Folder for prompt engineering
      ├── data.py
      └── prompt.py
  ├── /nlp                   # Folder for nlp processing
      ├── models.py
      └── nlp.py
  ├── /voice                 # Folder for voice processing
      └── voicetotext.py
  ├── telegram_bot.py        # Main script for the Telegram bot
  ├── update_data.py         # Script to update the data
  ├── utils.py               # Some utils functions
  ├── config.json            # Configuration file
  ├── requirements.txt       # Project dependencies
  ├── LICENSE                # License
  └── README.md              # This documentation file

---

🧑‍🤝‍🧑 Contributing

1. Fork the repository.
2. Create a branch (git checkout -b new-feature).
3. Make your changes and commit (git commit -am 'Add new feature').
4. Push your changes (git push origin new-feature).
5. Open a pull request.

---

📄 License

This project is licensed under the GPL License. See the LICENSE file for details.