Replies: 1 comment
-
being not a bug since what you are asking is fully dependent to the model you are using which we are not the authors or trainers of it, so you can train a new model for your specific language in XTTSv2 for now or other TTS engines (VITS?, FAIRSEQ?, TACOTRON2?, BARK?, YOURTTS?). also you don't provide which language you are talking about, with a text exemple etc... open an issue with 5 lines of text without log, audio example, etc... is not helpful. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Script Mode
Script Version
Process Mode
Operating System:
Command Line used to start ebook2audiobook
Describe the bug
I've noticed that when a text contains a font with diacritics the letters with a diacritic are ignored by the voice and so the word 'mettā' gets pronouced 'mett 'which makes it slightly unintelligible - it would be better if the letters with diacritics were either pronounced correctly ie a line over a letter a is a long a sound, or the letter with diacritic is converted to a letter without a diacritic and so the word would be pronounced 'metta', which is intelligible even if not pronounced strictly correctly.
Terminal log
n/a
Beta Was this translation helpful? Give feedback.
All reactions