GUIDE

InShot text to speech: AI Speech and other voices

InShot is one of the most popular phone video editors for Reels, TikTok and Shorts, and many people ask whether it can read text aloud. It can: the app lists AI Speech, which turns text into speech with a choice of voices.

This guide explains what AI Speech in InShot offers and how to bring in voices InShot does not have, such as game and cartoon characters or a narrator in your language, made with SuperClonica.

AI Speech in InShot

According to its App Store description, InShot has AI Speech, which turns text into natural-sounding speech with AI text to speech and an AI voice generator. An update added editing of the speech’s speed and emotion. Update the app and look for AI Speech among the editing tools; menu names change between versions, and the listing does not say whether the feature needs InShot Pro.

When you want a different voice

Built-in voices are shared by everyone who uses the app, so videos start to sound alike. For a character voice, a deep trailer narrator or a voice that matches your channel, make the audio elsewhere and add it to InShot as your own sound.

A short line from your phone

On the page of any SuperClonica voice, paste 30 to 100 characters, choose the language and press Voice it. In a few seconds you can hear the line and download it, with no account. That is enough for a hook, a punchline or a meme reply; add the file in InShot through its music and sound tools.

A whole script on your computer

For longer voiceovers, install the free SuperClonica engine on a computer with python -m pip install superclonica-cpu, save the script as a text file and run superclonica --text-file script.txt --voice voice.json --lang en --normalize --output voiceover.mp3. It runs offline on an ordinary processor with no character limit, and --normalize keeps every clip at the same loudness. Send the MP3 to your phone and add it to the InShot project.

Fair use of character voices

Character voices in the catalog are fan-made. Use them for parody, memes and fan videos, label them when people could mistake them for the original, and never make a real person appear to say something they did not.