Anki text to speech: audio for every card
Hearing a word every time you review it makes it stick, but recording a deck of a thousand cards is not realistic, and add-ons that call online services depend on them staying available. A deck with audio generated once, offline, keeps working forever.
The free SuperClonica engine voices every card of a deck on your own computer, in natural voices and 31 languages. This guide shows how to export the cards, voice them in one run and attach the audio with Anki’s sound tags.
What you need
Most catalog voices were recorded in Russian and keep a slight accent in other languages, so test the voice on a few of your words first.
- Anki on your computer and a deck to give a voice to.
- The free SuperClonica engine and Python 3.10 or newer.
- A voice that suits the language you study: a stock voice or one recorded in that language.
- No graphics card: an ordinary processor is enough.
Hear a card free online
On the page of any voice, paste a typical card sentence of 30 to 100 characters, choose the language and press Voice it. In a few seconds you hear how your deck would sound, with no account.
Export the cards
Single words sometimes sound better inside a short phrase; voice the example sentence if your card has one.
- Export the notes as plain text from Anki, or keep your source spreadsheet.
- Give every row a unique id, such as card_0412, and keep the text to voice in its own column.
- Save the table as CSV in UTF-8, so accents and non-Latin scripts survive.
- Write abbreviations and numbers the way they should be heard.
Voice the whole deck offline
Install the engine with python -m pip install superclonica-cpu. In Python, after import csv, from superclonica import Superclonica and tts = Superclonica(), one loop voices the whole deck: for row in csv.DictReader(open("deck.csv", encoding="utf-8")): audio, rate = tts.synthesize(row["front"], voice="F2", lang="es", seed=7); tts.save(row["id"] + ".ogg", audio). OGG files are small and play in Anki on every platform. Use --speed on the command line, or voice slower takes separately, if you want a learner-friendly pace.
Attach the audio in Anki
Copy the audio files into the collection.media folder of your Anki profile. Add a field with the sound tag of each card, such as [sound:card_0412.ogg], to your CSV and import it back into Anki, or paste the tags into an audio field. Anki plays the file when the card is shown, offline, on desktop and on phones after syncing. Re-voice a single card any time by running the loop for that row with the same seed.