GUIDE

Text to speech for presentations: narrate your slides

A narrated presentation can be watched without you: a product demo, a training course, a report for colleagues in another time zone. Text to speech turns your speaker notes into a calm, even voice-over, and you can fix a slide by editing a sentence instead of re-recording.

SuperClonica has neutral voices made for explanations, a free online test and a free engine that narrates the whole deck on your computer. This guide shows which voices suit slides, how to write a script for the ear, and how to voice every slide offline.

Browse voices Install the SuperClonica engine

Voices for presentations

All of them speak the 31 languages of Supertonic 3. Most catalog voices were recorded in Russian and keep a slight accent in other languages, so voice one slide in your language before you choose.

  • Business Male Voice: low, moderate intonation, made for presentations, instructions and explaining tasks.
  • Pleasant Announcer: low and restrained, for instructions and voice-over.
  • Young Analyst: mid-low with light emphasis, reads comparisons clearly.
  • Eva and Nika: neutral, even female voices for an address or an explanation.
  • Maya: a conversational female voice, free.
  • The ten stock voices of the model, M1 to M5 and F1 to F5: free in the engine.

Test a slide free online

On the page of any voice, paste a sentence from your notes, 30 to 100 characters, choose the language and press Voice it. In a few seconds you hear how the slide will sound and can download the recording. No account is needed.

Write the script for the ear

Read the script aloud once yourself: what you stumble over, a listener will too.

  • One idea per sentence; long sentences with nested clauses are hard to follow by ear.
  • Write abbreviations and units the way they should be read: “kilometers”, not “km”.
  • Figures can stay as digits: the engine spells them out in 27 languages. Check sentences with numbers in Hindi, Greek, Estonian and Croatian, where it does not.
  • End each slide’s text with a full stop so the voice finishes calmly.

Narrate the whole deck offline

Install the free SuperClonica engine: python -m pip install superclonica-cpu. Save each slide’s notes as a text file and voice it with one command: superclonica --text-file slide-01.txt --voice F2 --lang en --normalize --output slide-01.wav. The --normalize option brings every slide to the same loudness, −18 LUFS. The result is a 16-bit WAV file that you insert on the slide as audio in PowerPoint or another editor. It runs offline on an ordinary processor, with no character limit, so confidential decks never leave your computer.

Your company’s own voice

For a company, one recognizable voice across all decks and courses is worth more than any stock voice. With a colleague’s consent, record a clean phrase of 5 to 10 seconds, create a voice in the studio and train it on a computer with an NVIDIA graphics card; training on your own hardware is free. Without a suitable card, order the voice on the exchange from an author who trains it.