Mandarin auto-generate audio is terrible

I used to use the auto-generate audio option in the past, and even though the voices were always too robotic and artificial, it was still usable.
But lately it is terrible, because whenever the audio is generated, the words have pauses in between them that make it even worse to listen to.
It didn’t use to be like this.
Will it be fixed in the future?

Something else on this topic: Will LingQ ever implement the newer synthetic voices on the market that sound more like human voices?

Thank you.

3 Likes

You can use my extension script for realistic AI-TTS.


This is a demo using the TTS. Much better than existing mechanical one:

I’m excited to try. I’ll give it a day or two using it and I’ll tell you what I think about it :smiley:

Thanks!

1 Like

How can I use your script/add-on to generate the audio for the lesson without having to copy the lesson to Gemini/Chat GPT Interface?

https://aistudio.google.com/generate-speech

Thanks!
I find your extension to be quite useful
Let me ask you for something that could be a life-saver for most of us (if only LingQ itself would implement it would be great both ways)

LingQ has an option to PRINT OUT a lesson.
Is it possible that you include the created LingQs for that lesson in a word list with some spaces to practice the word?

Even better, an option to have the full text + the text with blanks where the LingQs would be and then a word list so we can fill in the blanks?

Hope you like my idea and hope it’s not too difficult to code as well.

(fingers crossed)

1 Like

Getting the words in a lesson is quite easy. And coding will not take too much time. But I’m working on another project for now. If I have some time, I will add the feature to my add-on and let you know :slight_smile:


Please answer this question: