🎙️ Voice Assistant Documentation 🎙️

🚀 This project allows you to create a sophisticated voice assistant using the OpenAI API and the Google speech recognition module. The code is divided into three main modules for a robust and clear code structure.

⚠️ Prerequisites: Before running the project, make sure to add the OPENAI_API_KEY environment variable with your personal OpenAI API key and to have VLC media player installed on your system. Also, make sure you have Python and pip installed in your environment.⚠️

🛠️ Local Installation

To install the project locally, follow these steps:

1. Create a virtual environment using venv:

python3 -m venv venv

2. Activate the virtual environment:

   • On Windows:

   venv\Scripts\activate

   • On Unix or MacOS:

   source venv/bin/activate

If you're using Linux, run the script_linux.sh script:

 ./script_linux.sh

and run

pip3 install -r requirements_linux.txt

If you're using MACOS:

3. Install project dependencies using pip:

pip3 install -r requirements.txt

👂 Module: voice_recognition.py

The voice_recognition.py module acts as the "ears" of our voice assistant. This module handles listening to user input and its subsequent conversion to text.

🎯 This module includes the VoiceRecognition class with methods:

• listen() 🎤: Listen to user input via the microphone.
• decode_speech(audio) 📝: Convert voice input to text.

💬 Module: dialogue_management.py

The dialogue_management.py module is the brain of our assistant, managing the dialogue between the user and the assistant.

🎯 This module contains the DialogueManagement class with methods like:

• _init_dialogue() 💼: Initialize dialogue by setting up initial system statements.
• add_dialogue(role, text) 🗣️: Add new dialogues to ongoing discussions.
• chat_completion() 🤖: Generate an appropriate response using the OpenAI API.

🗣️ Module: voice_assistant.py

The voice_assistant.py module is the main module and entry point of this project. It combines the dialogue_management.py and voice_recognition.py modules into a single voice assistant application.

The VoiceAssistant class in this module has the following methods:

• escape_character(text) 🧹: Remove escape characters from text, making input safe.
• run() 🏃‍♂️: Start the interaction with the voice assistant and continue until interrupted.

---

💡 Available Text-to-Speech Services:

• PYTTSX3: Pyttsx3
• OPENAI: OpenAI
• GTTS: gTTS
• FUN_VOICE: Fun Voice
• ELEVENLABS: ElevenLabs (requires key configuration)

ℹ️ Next Implementation: Integration with Home Assistant will soon be added for enhanced automation and home control.

💡 To run, execute:

python3 main.py

💡 To change the Text-to-Speech service, modify the value of the service variable in the __init__ method of the VoiceAssistant module.

• TextToSpeechService(service=ServiceType.GTTS) 🗣️

Have fun! 🎉

---