Canva text to speech: voice-overs for videos and presentations
Canva makes it easy to design a video, a presentation or a social post, and a voice-over turns it into something people watch to the end. Canva itself does not read text aloud; voices come from apps or from an audio file you upload.
SuperClonica gives you that audio file: natural narrators and character voices, free to try online, and a free engine for whole scripts. This guide shows how to voice the text, which files Canva accepts and how to put the voice where it belongs in your design.
How voice-overs work in Canva
A file you made yourself stays the same wherever you reuse it, in Canva or in any other editor.
- Canva accepts uploaded audio in M4A, MP3, OGG, WAV and WEBM files of up to 250 MB.
- Text-to-speech inside Canva comes from apps, each with its own voices, limits and terms.
- An uploaded voice-over works in videos, presentations and animated social posts alike.
- SuperClonica voices speak 31 languages; Chinese is not supported yet.
Voice a line free online
On the page of any voice, type a line of 30 to 100 characters, choose the language and press Voice it. In a few seconds you can listen and download the recording, with no account. A short line for an intro or a social post fits easily.
Add the voice to your design
For a design with many pages, make one file per page: a change on one slide then means voicing just that slide again.
- Voice the text and download the recording.
- In Canva, open Uploads and upload the file, or drag it straight into the editor.
- Drag the audio onto the timeline and trim it to the pages it belongs to.
- Lower the music under the voice so every word stays clear.
A voice-over for every slide
For a longer presentation or a course, install the free SuperClonica engine on a computer: python -m pip install superclonica-cpu. Save the notes of each slide as a text file and voice it with one command: superclonica --text-file slide03.txt --voice F3 --lang en --normalize --output slide03.mp3. --normalize keeps every slide at the same loudness, and the engine runs offline on an ordinary processor with no character limit. The ten stock voices are free; catalog voices are bought once.
Voices and rights
For a brand, a client or a course you sell, use the ten stock voices of the model, M1 to M5 and F1 to F5, or your own voice trained in the SuperClonica studio from a recording of a speaker who agreed. Character voices in the catalog are fan-made and not endorsed by the rights holders: keep them for parody and fan content. Tell viewers when a voice is generated if it could be mistaken for a real person.