Re:Show HN: Text-to-speech and speech-to-text open-source so...
by lunixbochs on Tuesday January 28, 2020 @10:22AM
(#22170173)
How does accuracy compare to what you’re using now?
[ Parent
| on HN ]
Re:Show HN: Text-to-speech and speech-to-text open-source so...
by ftreml on Tuesday January 28, 2020 @10:52AM
(#22170455)
the german model is from the kaldi tuda recipe with WER of 15%.
the english is from the tedlium recipe with WER of 7%.
room for improvement, but for our original purpose it was sufficient.