AI Storytelling Avatar Clone

I want to bring a lifelike, AI-powered “digital human” to life—one that exists purely to entertain by telling engaging stories on request. Picture a personable avatar (voice-only or full-body video, whichever your toolkit handles best) that listens to a prompt, spins an original tale, and delivers it with natural pacing, expression, and emotion.

Here’s what I need the finished clone to do:
• Accept text or voice prompts from the user.
• Generate or select appropriate stories, then narrate them fluidly in real time.
• Maintain a consistent, memorable persona so the audience feels they’re listening to the same storyteller every time.

A smooth integration of generative-text models (e.g., GPT-based), voice-cloning or TTS engines, and—if you offer it—avatar animation tech would be ideal. I’m open to your preferred stack as long as the end result is convincing, easy for me to run, and requires minimal manual intervention once deployed.

Please outline:
• The tools/libraries you plan to use.
• Any hosting or runtime requirements.
• A brief timeline to reach a working demo and then a polished release.

If you have samples of past AI voice or avatar work, include them—they’ll help me decide quickly.