Turn your voice into an illustrated story.
Just tell a story out loud. NarraFlow listens, understands who and where and what happens, and paints each scene as you go. Ask it to change anything, by voice, and watch the pictures follow.
Voice is powered by AssemblyAI's Voice Agent API and needs a microphone and a secure (https) connection. The demo replays a short original story with no microphone and no sign-in.
Say it. See it. Steer it.
A conversation with a co-storyteller, not a record button.
Tell your story
Talk naturally, with pauses and false starts. “Once upon a time, Amara lived under the mountains…” NarraFlow waits while you think.
Watch it appear
Characters, places and scenes are understood as you speak. Every scene is illustrated, and each character keeps the same look throughout.
Steer by voice
“Make the tree blue.” “Change her dress to green.” “Regenerate scene two.” Then say “play the story” to watch it with narration and captions.
How it works
The voice agent doesn't just talk. It acts on a structured story through tool calls.
- Real conversation. Turn detection tuned for storytellers who pause, and instant barge-in when you talk over the agent.
- Tools, not just talk. Create, add and edit scenes, change a character everywhere, restyle the whole story, reorder or remove scenes (with a spoken confirmation), play and export.
- Consistent characters. Each character's look is stored once and injected into every illustration prompt.
- Never blocks the conversation. Painting runs in the background; the agent keeps talking while scenes appear.
- Secure by design. Server-issued temporary tokens, rate-limited, no secrets in the browser.
Who it's for
Anyone who tells stories better than they draw them.