-
I've tried to train with my own non-English dataset (~ 3 hours, every wav is from 5 to 8 seconds) but the alignment seems very strange.
 Doesn't work with [0, 1 $1(, requires [0 0, 1 $1( instead.
2) Output envelope is different from [vline~] and can substantially change the sound when very fast envelopes are required for things lik…
-
# Text-to-Speech Synthesis
Text-to-Speech is a speech generation task that converts written language into its spoken form.
## Task Objective
Text-to-Speech Synthesis (TTS) is an essential ta…
-
Taken from not-yet-done list on #3 :
- [x] Reading input from files
- [x] 12 bit value output to lists `:` (should be easy, just haven't done it)
- [ ] the `' ... '` construct, which is termed a …
hornc updated
6 months ago
-
Window.js should have a sound API.
I'm not sure whether adopting the entire HTML5 Web Audio API is the way to go:
https://developer.mozilla.org/en-US/docs/Web/API/Web_Audio_API
This has a lot…
-
@rafaelvalle @pravn @ksaidin @CookiePPP
I have American English accents female 7.4 hours of audio dataset. I've removed start and end silence using @Yeongtae [yeongtae's processing](https://github…
-
Dear Marco,
Wishing you a happy new year.
I've been working extensively with MelGAN-VC in my practice the better part of last year.
I have recently run into the issue that the model is a bit …
-
Hi y’all
after playing the new re-release of doom and experiencing the DMXOPL stuff they added into it it struck me that it would fit in perfectly with the overall feel and aesthetic retro goes for…
-
(From http://rpmlint.zarb.org/cgi-bin/trac.cgi/ticket/45, pkarlsen@…)
As synthesis doesn't store package filenames, this will lead to difficulties installing these.
Changed 7 years ago by scop
sta…