GUIDE

Google Docs text to speech: listen to a document or save it as MP3

Google Docs has no button that reads a document aloud. Listening to a doc is really two different jobs: having the text read while you follow it on screen, and getting an audio file you can play anywhere, on a walk or in the car.

For the first, the tools of your device do the job. For the second you need a text-to-speech engine. SuperClonica gives you natural voices in 31 languages, free to try online, and a free engine that turns a whole document into an MP3 on your own computer, so the text never leaves it.

Browse voices Install the SuperClonica engine

Listening inside Google Docs

These tools read with your system’s voice and only while the document is open; for a natural voice you can take with you, make an audio file.

  • Google Docs has no read-aloud button of its own; it supports screen readers, which you turn on in its accessibility settings.
  • On a Chromebook, Select-to-speak reads any text you select, documents included.
  • On a Mac, Spoken Content can speak a selection; on Windows, Narrator reads the screen.
  • Add-ons from the Google Workspace Marketplace put a read-aloud sidebar inside Docs.

Try a natural voice free online

On the page of any voice, paste a sentence from your document, 30 to 100 characters, choose the language and press Voice it. Compare a few voices this way and keep the one that is easiest to listen to for a long time.

Turn a document into an MP3

Digits are read as plain numbers, so write dates and years as words when they matter.

  • Download the document as plain text: the download options in the File menu include a .txt file.
  • Install the free engine: python -m pip install superclonica-cpu.
  • Voice the whole file into one MP3: superclonica --text-file document.txt --voice F2 --lang en --normalize --output document.mp3.
  • Check the .txt first: tables and footnotes may need tidying, and web addresses are read out word by word, so delete them.

Listen on the go

Put the MP3 into Google Drive and play it in the Drive app on your phone, or copy it to any music player. The engine has no character limit: long text is split into pieces and joined into one file, and --normalize keeps the loudness even at −18 LUFS. The free online test voices 30 to 100 characters at a time, handy for a heading or a quote, not for a whole document.

Voices for what you share

For audio you share, such as class materials, a newsletter or a company document, use the ten stock voices of the model, M1 to M5 and F1 to F5, free under the OpenRAIL-M license, or your own voice trained in the SuperClonica studio. For private documents, the engine is the safe choice: it runs offline.