# machine-learning
- u
(Dawn) Will Draw Fictional Women
12/14/2021, 4:12 PMaverage uberduck user - h
haru0l
12/14/2021, 4:14 PMaverage discord user - z
Zachary(retired f/ voice clonin)
12/14/2021, 4:14 PMaverage scammer - h
haru0l
12/14/2021, 4:15 PMaverage - p
Pekka
12/14/2021, 10:55 PMIs there a software that can transcribe individual audio files and rename the files with the transcribed audio. - u
(Dawn) Will Draw Fictional Women
12/15/2021, 5:37 AMkind of but its only available to contributors - u
user
12/15/2021, 6:14 AMAI generated Alaskan Malamute in Bangkok - h
Hex
12/15/2021, 11:02 AMHai! I want to make a model, I have understand how AI works, and the more basic stuff. Could anyone point me to a post here or something where it's explained how to make a model for uberduck? Should I just download tacotron2 set it up then watch a tutorial about where to put the voice samples and their subtitles? Or is there more to it? - h
hecko
12/15/2021, 11:10 AMyou don't have to download it actually, all the training will happen on google's servers - h
hecko
12/15/2021, 11:10 AMsee the tutorial in #841422801965416538, it's kind of poorly written and the links are outdated (use the ones in #841437191073955920 instead) but it works in a pinch - h
Hex
12/15/2021, 11:11 AMOkay! Thanks! - u
🇵🇷TehPuertoRicanSpartan🇵🇷
12/15/2021, 2:57 PMwhat ai did you use? - e
EdFan Sus (Team Fanatical)
12/15/2021, 8:00 PMLiterally movies be like - g
Gushiavo_N2
12/16/2021, 5:46 PMwhat think for Batch size next for 44 wavs? - h
hecko
12/17/2021, 12:10 AM22 - t
The Wiggles and High School D×D
12/17/2021, 12:53 AM@User Okay. I think I'm just going to use that notebook you linked me to redo the MLP process (in that I'll have all the MLP voices this time) for the sake of simplicity on my part and probably a better model on my end. - t
The Wiggles and High School D×D
12/17/2021, 12:55 AMI'll have to run the encoder step again but I guess I don't mind - m
mega b
12/17/2021, 2:33 AM4-8 might be a good range - t
The Wiggles and High School D×D
12/17/2021, 2:57 AMI got a tricky proposition I can't figure out - t
The Wiggles and High School D×D
12/17/2021, 2:58 AMI have this audio file of a conversation between two people - t
The Wiggles and High School D×D
12/17/2021, 2:58 AMThat I want to deepfake - t
The Wiggles and High School D×D
12/17/2021, 2:59 AMWhat I need to do is find some kind of video to model on two deepfakes (one for each person) - t
The Wiggles and High School D×D
12/17/2021, 3:00 AMI then need to somehow integrate both the audio file and the faces onto these deepfakes - t
The Wiggles and High School D×D
12/17/2021, 3:00 AMOr, if you will - t
The Wiggles and High School D×D
12/17/2021, 3:00 AMMake the template lip sync to my audio - t
The Wiggles and High School D×D
12/17/2021, 3:00 AMAnd with a program like, say, yanderify, I would use that video for my people's faces - t
The Wiggles and High School D×D
12/17/2021, 3:01 AMI can't figure out how to approach this - h
haru0l
12/17/2021, 11:10 AMyou probably have to separate them - h
haru0l
12/17/2021, 11:10 AMcrop it out - j
Justin
12/17/2021, 2:40 PMruDALL-E Outpainting result: (Input img 👇 )