Give your character a voice: a wizard, knight, pirate, fairy, villain or giant. Type the line, download the MP3.
Your result will appear here.
Every story needs voices. Pick an archetype (an old wizard, a young knight, a pirate captain, a tiny fairy, a gruff blacksmith, a villain, a robot helper, a cowboy or a giant), choose a mood and type the line. Our AI voice model performs it in character, and you can download the clip for a game, an animation, a tabletop session, a story video or a school play.
Bridge troll riddle
Wizard advice
Knight oath
Pirate orders
Fairy giggle
Blacksmith
Villain reveal
Robot helper
Mix any character with any mood: an angry fairy, a sad giant, a mischievous knight. The mood changes how the line is delivered, so the same words can sound brave, scared or sneaky. Every press records a new take, which makes it easy to audition a few readings before you pick one.
The voices are designed archetypes, made for your own creations. They do not copy actors or characters from films, shows, books or games, and the tool refuses requests to copy a real person's voice. Give your character a name and a personality in the line itself, and it becomes yours.
Write lines the way the character would talk: short and gruff for a blacksmith, poetic for a wizard, slangy for a pirate. Use one clip per line so you can trigger them separately in a game engine or video editor. Keep a list of the character and mood you used so every line of the same character matches.
Need a specific type? The Old Wizard Voice Generator, Robot Voice Generator, Anime Voice Generator and Scary Voice Generator each have more voices of one kind.
Yes, it is free to use, and each clip downloads as an MP3.
Yes, you can download the clips and use them in your own projects.
No. It makes original archetype voices only. Requests to copy real people or existing characters are refused.
Use the same character and mood for every line. Each take varies a little, so record a few and keep the ones that match best.
Up to 250 characters, about 20 seconds of speech per clip.