MagicVoice 1.5
MagicVoice 1.5 is MagicShot's own speech model. It turns written scripts into natural voiceovers in 29 languages, with control over pacing and tone through punctuation, write it like it's spoken and the delivery follows. It powers Text to Speech and the audio layer across MagicShot's video tools.

What is MagicVoice 1.5?
What MagicVoice 1.5 can do
Voiceover in 29 languages
Turn a written script into natural spoken audio across 29 languages from one model.
Direct delivery with punctuation
Control pacing and tone by writing the script the way it should be spoken. The delivery follows your punctuation.
Clone a specific voice
Recreate a particular voice from a short reference sample for consistent narration across a series.
How to use MagicVoice 1.5 on MagicShot
Open a tool
Head to Text to Speech or any audio tool. No model settings to configure.
Add your prompt
Describe what you want, or upload your own media to work from. Plain language is all it takes.
Generate and download
MagicVoice 1.5 returns your voice clip in seconds, ready to download and use commercially on paid plans.
Tools powered by MagicVoice 1.5
Text to Speech
Paste a script, pick a voice, narration that doesn't sound like a robot reading.
One subscription. Every top audio model.
MagicVoice 1.5 is one of the AI audio models built into MagicShot. Instead of paying for, learning, and switching between a different app for each one, you get them all in a single place. We benchmark the newest models as they ship and swap in the best, so your results keep improving and everything you make is yours to use commercially.
Other AI audio models on MagicShot
Lyria 3.5
FeaturedGoogle DeepMind's Lyria 3.5 music model, built for full songs with vocals that follow your lyrics. On MagicShot it is offered in the Music Generator.
View Lyria 3.5 →MagicVoice 1.5 FAQ
Ready to create something magical?
Make images, videos and voiceovers in seconds.