Is it possible to override or replace incorrect TTS?

I use Anki to learn French, and sometimes the TTS pronunciation is incorrect. Is there a way to override TTS within a specific note, so you can replace it with a manually recorded version?

You could duplicate your notetype, delete the tts code and using that notetype for your selected notes.

1 Like

That’s a decent idea, thanks!

Ideally there’d be a way of having just one note type with an audio file field, with the note knowing to override the TTS if there’s an audio file in the audio field. Is that possible (or something equally straightforward)?

Another inelegant (but faster) solution while someone else provides a better solution, would be to use two French fields in your notetype, with and without TTS. You would just need to edit > cut & paste the content between them in those cards using Mp3.

By the way, I also use Anki to learn French, Microsoft TTS voices in Windows 10, and (while not perfect) I find the pronunciation pretty usable. What voices are you using? Could you give an example of incorrect prinunciation?

You can use Conditional Replacement. Try the following:

{{Audio}}
{{^Audio}}
    {{tts fr_FR:Front}}
{{/Audio}}

If you put [sound:foo.mp3] in the Audio field, that audio file will be used, and if you leave the field empty, TTS will be used.

2 Likes

Perfect – thank you!

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.