Ongezellig
IAZ and soft NAZ
ZWABAG
[Return] [Thread list] [Index] [Catalog] [Bottom] | Replies: 10 | Views: 2 | Currently viewing: 1
File:
1784499431489.webm
(5.70 MB, 358x360, Running-in-the-60s-Liedje-van-C-_Ongezellig-fan-song_.webm)
I cordially invite all Zaryans to listen to the song by Running in the 60's
File:
1784499431489.mp3
(5.35 MB, loser_babies.mp3)
>>16275
Oh hey its you from the 4cuck thread
Oh hey its you from the 4cuck thread
File:
1784499431489.mp3
(10.70 MB, 1707040078002-1.mp3)
>>16280
Yup
Yup
>>16288
I use either these two sites for AI covers
Applio: https://huggingface.co/spaces/IAHispano/Applio
CoverGen_NO_UI colab: https://colab.research.google.com/drive/1u1brjK8IZt647UsbZuGYfW29oFM2I4tk
I mostly use CoverGen since Applio can take forever to output if a bunch of users are using it all at once, but I think Applio can produce better results compared to CoverGen when you get the chance.
I don't know where you're getting the models for the girls' voices, but this is where they are all hosted: https://huggingface.co/filler5736
As far as making it sound better, that mainly falls on the dataset used and messing around with the output settings like the Index Rate/Search Feature Ratio (how much little details are added to the output that the original voice has, but may produce more robotic consonants) or what pitch extration algorithm was used (ie. mangio-crepe or rmvpe).
Anything else is being a bit knowledgable on audio/music production so things like EQ/Compression/Reverb/etc. could help make the AI voices sound less shite.
I use either these two sites for AI covers
Applio: https://huggingface.co/spaces/IAHispano/Applio
CoverGen_NO_UI colab: https://colab.research.google.com/drive/1u1brjK8IZt647UsbZuGYfW29oFM2I4tk
I mostly use CoverGen since Applio can take forever to output if a bunch of users are using it all at once, but I think Applio can produce better results compared to CoverGen when you get the chance.
I don't know where you're getting the models for the girls' voices, but this is where they are all hosted: https://huggingface.co/filler5736
As far as making it sound better, that mainly falls on the dataset used and messing around with the output settings like the Index Rate/Search Feature Ratio (how much little details are added to the output that the original voice has, but may produce more robotic consonants) or what pitch extration algorithm was used (ie. mangio-crepe or rmvpe).
Anything else is being a bit knowledgable on audio/music production so things like EQ/Compression/Reverb/etc. could help make the AI voices sound less shite.
>>16523
Little extra tip: If you find that the voice jumps around way too much between pitches using rmvpe, trying switching to crepe/mangio-crepe and set the hop length really high (like ~450). This will make it so that the jump between pitches will be much slower and won't be so drastic. You can then cut out jumpy parts from the rmvpe voice output and splice in ones from the crepe/mangio one. And if you know or have something like Melodyne or NewTone from FL Studio, that could definitely help with fixing pitch issues if needed, but it is a bit tedious.
I've also got this Google Doc that details the best tools for things like vocal/instrumental isolations currently available; updated frequently: https://docs.google.com/document/d/17fjNvJzj8ZGSer7c7OFe_CNfUKbAxEh_OBv94ZdRG5c/edit#heading=h.roiuj54hzww3
But mainly I've been going to this site for my voice/instru isolations for the sake of convenience: https://mvsep.com/en/home
(Requires an account, but I'm sure you're smart enough to know how to use burner emails)
Little extra tip: If you find that the voice jumps around way too much between pitches using rmvpe, trying switching to crepe/mangio-crepe and set the hop length really high (like ~450). This will make it so that the jump between pitches will be much slower and won't be so drastic. You can then cut out jumpy parts from the rmvpe voice output and splice in ones from the crepe/mangio one. And if you know or have something like Melodyne or NewTone from FL Studio, that could definitely help with fixing pitch issues if needed, but it is a bit tedious.
I've also got this Google Doc that details the best tools for things like vocal/instrumental isolations currently available; updated frequently: https://docs.google.com/document/d/17fjNvJzj8ZGSer7c7OFe_CNfUKbAxEh_OBv94ZdRG5c/edit#heading=h.roiuj54hzww3
But mainly I've been going to this site for my voice/instru isolations for the sake of convenience: https://mvsep.com/en/home
(Requires an account, but I'm sure you're smart enough to know how to use burner emails)
Post actions
[Return] [Thread list] [Index] [Catalog] [Top] | Replies: 10 | Views: 2 | Currently viewing: 1