Closed tom-huntington closed 2 years ago
I followed the install from source instructions and it produced a different tts audio which gave the correct results
f000001 0.000 5.880 "From fairest creatures we desire increase," f000002 5.880 9.240 "That thereby beauty’s rose might never die," f000003 9.240 11.920 "But as the riper should by time decease," f000004 11.920 15.280 "His tender heir might bear his memory:" f000005 15.280 18.880 "But thou contracted to thine own bright eyes," f000006 18.880 22.760 "Feed’st thy light’s flame with self-substantial fuel," f000007 22.760 25.680 "Making a famine where abundance lies," f000008 25.680 31.240 "Thy self thy foe, to thy sweet self too cruel:" f000009 31.240 34.440 "Thou that art now the world’s fresh ornament," f000010 34.440 36.960 "And only herald to the gaudy spring," f000011 36.960 40.680 "Within thine own bud buriest thy content," f000012 40.680 44.480 "And tender churl mak’st waste in niggarding:" f000013 44.480 48.080 "Pity the world, or else this glutton be," f000014 48.080 53.240 "To eat the world’s due, by the grave and thee."
I was getting the same problem again, and I had to run
python setup.py build_ext --inplace
again to fix it
On windows I got the correct alignment but not on WSL. I'll try and solve this myself, just posting in case it helps someone else.
Audio produced by espeak: https://user-images.githubusercontent.com/55266932/170434589-970b0d03-1a25-4f3a-91fd-b615e0cc8e7a.mp4