Bedtime story AI voice: a calm narrator for fairy tales
A bedtime story needs a calm, warm voice and a few characters: a sleepy bear, a brave little rabbit, a kind old wizard. An AI voice can read the story you wrote or a fairy tale in the public domain, as many times as the child asks, while the characters keep their own voices.
SuperClonica has calm narrator voices, character voices for children’s stories, a free online test and a free engine that reads whole stories on your computer. This guide shows which voices suit bedtime, how to write a story for the ear and how to make an audio fairy tale offline.
Storyteller voices
Keep one narrator for every story so the child recognizes the voice, and give each character a voice of its own.
- Anna: a calm, fairly low female voice; stories can be read simply, with no announcer tone.
- Sofia and Nika: even, pleasant female voices for a narrator.
- Keeper of the Light: a kind old wizard, for legends and lessons.
- Little Boy and Schoolgirl: child voices for the characters.
- Valera the Gummy Bear: a high, bouncy cartoon voice for the funny character. Free.
- Maya: a conversational female voice, free.
Try a line free online
On the page of any voice, paste the first line of your story, 30 to 100 characters, choose the language and press Voice it. In a few seconds you can listen and download the recording. No account is needed.
Write a story for the ear
For bedtime, a slightly slower narrator helps: in the engine, try a speed of 0.95 instead of the default 1.05.
- Short sentences and simple words, the way you would tell it aloud.
- Refrains that repeat: children love to hear the same line again.
- Characters’ lines on separate lines, so you can voice them with their own voices.
- A quiet ending: the last sentences slower and softer, without exclamation marks.
A whole fairy tale offline
Install the free SuperClonica engine: python -m pip install superclonica-cpu. Save the narrator’s text as a file and voice it with one command: superclonica --text-file story.txt --voice anna.json --lang en --speed 0.95 --normalize --output story.wav. Voice the characters’ lines the same way with their voices and put everything together in any audio editor. The engine runs offline on an ordinary processor, with no character limit, and the 16-bit WAV files play on any phone, tablet or speaker without the internet.
A parent’s own voice
The voice a child loves most is a parent’s or a grandparent’s. With their consent, record a clean phrase of 5 to 10 seconds, create a voice in the studio and train it on a computer with an NVIDIA graphics card; training on your own hardware is free. Then new stories can be read in a familiar voice even when the parent is away. Without a suitable card, order the voice on the exchange from an author who trains it.