This project combines ChatGPT with Moegoe TTS to create an interactive "Chatting Waifu" for the Marai chatbot platform. It targets users seeking a personalized, voice-enabled AI companion that can engage in conversations and respond with synthesized speech, enhancing the user experience for Marai bot operators.

How It Works

The system leverages a modular architecture, integrating ChatGPT for conversational AI, Moegoe TTS for speech synthesis, and miraicle's mirai-http-api for connecting to the Marai chatbot. It supports multi-character voice options and includes speech recognition for voice input, aiming for a natural and human-like interaction.

Quick Start & Requirements

• Install dependencies via pip install -r requirements.txt.
• Requires Python >= 3.7 (example uses 3.10).
• Download and place TTS models into the model directory.
• Obtain verify_key and QQ number for mirai-http-api.
• Obtain ChatGPT session token from browser developer tools.
• Run using python ChatWaifu_marai.py.
• Model download links provided via Google Drive and Aliyun Drive.

Highlighted Details

• ChatGPT conversational chat.
• Speech synthesis for responses.
• Support for multiple character voices.
• Speech recognition for voice input.
• Integration with Marai chatbot and Live2D.

Maintenance & Community

The project acknowledges dependencies on MoeGoe_GUI, TTSModels, and PyChatGPT. No specific community links or active maintenance signals are present in the README.

Licensing & Compatibility

The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

Currently, all users share a single conversation session due to ChatGPT limitations. The project appears to be a personal endeavor with limited community support or formal maintenance.