# machine-learning
- u
(Edd) bruh moment
07/05/2021, 9:28 AMWhat do you mean? - d
Den Co. Man
07/05/2021, 9:29 AMI hate amogus - u
user
07/05/2021, 10:21 AMtotally unfunny - o
OctolingVladTTS
07/05/2021, 1:22 PMNo problem 😉 - u
user
07/05/2021, 1:38 PMFunny enough me and @JabberHawk were planning to do Luigi from the cartoon but it's the super show. - j
JacobLenstar
07/05/2021, 1:41 PMnice - j
JacobLenstar
07/05/2021, 1:41 PMdanny wells - h
Hoodied
07/05/2021, 3:35 PMTrained an AI on a 100 audio clips from this - h
Hoodied
07/05/2021, 3:35 PMAn hour of high quality Peter Capaldi audio with no background audio this is a godsend - h
Hoodied
07/05/2021, 3:37 PMis this good??? I can't tell - h
Hoodied
07/05/2021, 3:37 PMit looks good - j
JacobLenstar
07/05/2021, 3:41 PMtry the synthesis colab - h
Hoodied
07/05/2021, 3:41 PMI'll try it in a tad after a few more Epochs are done - u
user
07/05/2021, 4:51 PMI tried to make a Warp Darkmatter voice with clips from Buzz Lightyear of Star Command: The Adventure Begins, but it didn't go over so well - i
i95s
07/05/2021, 5:43 PMyour so mad funny I forgot to laugh 😐 - h
Hoodied
07/05/2021, 5:51 PMhello, does anyone mind sending me a long blank transcript page so I don't have to keep pasting the wavs/number.wav|'s myself - d
Den Co. Man
07/05/2021, 6:00 PM@User here it is - d
Den Co. Man
07/05/2021, 6:00 PM200 lines long - h
Hoodied
07/05/2021, 6:00 PMthank you!! - d
Den Co. Man
07/05/2021, 6:00 PMglad to help - c
Clunkster
07/05/2021, 8:15 PMHow do I submit a reference audio? - s
Scramble
07/05/2021, 8:17 PMwrong channel and go to #859823464847638538 - c
Clunkster
07/05/2021, 8:17 PMOkay - c
CyberSage87
07/05/2021, 8:34 PMHere's what happened when I pulled Steven Merchant's other sources onto the mix. Interesting results. Its able to speak for longer, at least. - u
_n
07/05/2021, 8:35 PMoh my model wasn't all that bad, though i did some hifi-gan tuning - u
_n
07/05/2021, 8:36 PMit had some problems correctly pronouncing certain words, but it sounded halfway decent, if a bit robotic - u
_n
07/05/2021, 8:39 PMhere's my model i trained for 114 epochs: https://drive.google.com/file/d/10EHposlozZpkbfZ3JbfN3cbAteUo0vvz/view?usp=sharing here's the hifi-gan finetuned model: https://drive.google.com/drive/folders/1vbV9-qVXEYs0cQwRH3kRv-L-lzncek7w?usp=sharing - h
Hoodied
07/05/2021, 8:56 PMMy capaldi model is going well , but the Hifi gan speech seems to be messing up a lot - h
Hoodied
07/05/2021, 8:56 PMI'll train it a bit more