Using TikTok's text-to-speech voice well
TikTok's text-to-speech reads your on-screen text aloud in a synthetic voice. It is useful for faceless videos, quick tips and creators who prefer not to record their own voice, as long as the text is written for listening.
Write short, spoken sentences
Each text box becomes one spoken phrase, so keep boxes to one short sentence. Long boxes are read in a flat rush.
Use the words a person would say. Abbreviations, symbols and emoji are read literally or skipped.
Fix pronunciation
If a brand name or slang word comes out wrong, spell it phonetically in the text that is read aloud, and keep the correct spelling in a separate text box that is not voiced.
Numbers are read as numbers, so '2nd' or '3x' may sound odd. Write 'second' and 'three times'.
Control the pacing
Space text boxes so each one stays on screen until its line has finished. Overlapping boxes make voices talk over each other.
Add a short gap between key points. Silence for half a second makes the next line land.
Choose voice and language deliberately
Pick the voice that suits the tone and stick with it across a series, so returning viewers recognise it. Check the voice language matches the text; English voices reading Indonesian sound wrong, and vice versa.
Frequently asked questions
How long should each text box be for text-to-speech?
One short sentence per box, since each box becomes one spoken phrase. Long boxes get read in a flat rush.
How do I fix a word the text-to-speech voice mispronounces?
Spell it phonetically in the voiced text box and keep the correct spelling in a separate box that is not read aloud.
Why do text-to-speech voices talk over each other?
Because the text boxes overlap. Space them so each stays on screen until its line finishes, and add short gaps between key points.