🎬 Story Teller – Turn ideas into living stories
✍️ 1. Start with an idea
Everything begins with a single input:
your story concept.
A sentence…
or a detailed prompt…
Story Teller adapts to your creativity.
You’re not writing a script — you’re setting the direction.
👥 2. Define your characters (up to 4)
For each protagonist, you simply provide:
- Name
- Gender
- Age
- Optional personality or background notes
That’s it.
No dialogue writing.
No voice casting.
No technical setup.
🧠 3. AI-driven story & character intelligence
The QWEN-VL model takes over and does two critical things:
- ✨ It writes the full story based on your idea and characters
- 🎭 It designs a unique voice identity for each character (tone, age, personality, rhythm, emotional range…)
Each character becomes distinct — not just in text, but in voice.
🎨 4. Voices are created from scratch
Here’s where Story Teller truly stands apart.
Using QWEN-TTS VoiceDesign, each character’s voice is:
- Generated from description only
- Fully original
- Consistent across the entire story
⚠️ No reference audio
⚠️ No voice cloning
⚠️ No pre-recorded samples
The voices are born with the story.
🎭 5. Automatic multi-character dialogue
All generated voices are collected into a RoleBank, then passed to QWEN-TTS Dialogue Inference.
The system:
- Assigns the right voice to each line
- Handles pacing and pauses
- Produces a single, continuous audio narrative
What you get is not stitched audio —
it’s a coherent, immersive performance.
🔊 6. Final result: a complete audio story
At the end of the workflow, you have:
- A fully narrated story
- Multiple distinct characters
- Natural dialogue flow
- Ready-to-use audio output
Perfect for:
- 🎙️ Podcasts
- 📖 Audiobooks
- 🎧 Audio dramas
- 🎥 Story-driven videos
- 🧪 Creative experiments
🚀 Why Story Teller is different
- One workflow → full story
- No voice inputs required
- Fully AI-driven characters
- Modular, expandable, ComfyUI-native
- Designed for creators, not engineers