Name: Danielle (from NewJeans) | AI RVC Model
Brand: Weights
Rating: 4.2 (42 reviews)

WebKit

 Discover Danielle (from NewJeans) (RVC v2) 720 Epoch, the latest advancement in AI Voice Models powered by Weights' cutting-edge Retrieval-Based Voice Conversion (RVC) technology. This sophisticated model captures the unique vocal characteristics of K-pop sensation Danielle, allowing you to generate stunningly realistic AI music covers and text-to-speech applications. Carefully trained through a rigorous 720 epoch process using a high-quality 7-minute dataset and refined with a 40k sample rate, this RVC Model delivers unparalleled accuracy and authenticity. Ideal for musicians, producers, content creators, and voice artists alike, harness the power of our advanced AI tools to effortlessly create AI Covers or transform your texts into lifelike speech – all for free! Try it today and elevate your projects with the extraordinary talent of Danielle reimagined through state-of-the-art artificial intelligence.

Trained by me in RVC v2 ~~harvest method~~. Dataset is 7 minutes length. Retrained with 40k sample rate in mangio-crepe hop-16 method For those who want to use this model, please don't forget to give credits to: leejhay's digital voicebox on youtube *but be aware! recommended to not upload any copyrighted instrumental songs and its ai voice there as they will copyright strike your account. test it on your device locally* leejhay_stanpop on tiktok *same warning above*

Danielle (from NewJeans)

Danielle (from NewJeans) (RVC v2) 720 Epoch

leejhay

She sounds like Hanni but yeah I like that 😂 👍

Danielle (from NewJeans) (RVC v2) 300 Epoch (beta)

Danielle (from NewJeans) (RVC v2) 300 Epoch

with some editing skills, I can produced good quality of this ai cover. I hope you like this model ☺️

**New link updated! Huggingface is now standard for sharing voice models**

Sample of retrained voice model from 40k resampled dataset in crepe-hop 16 mangio-crepe. i've noticed that it can produce sound with less distorted breaths

is it ok to have this graph? that's the lowest value i could get. i don't think it's overtrained

Looks good. As long as you get the weight from the lowest point or close to it

mentosandrice

Lucifer Angel Satane Ghost

Trained from a most popular meme-able Kazakh news video clip that he speaks like a starting-up diesel engine, just almost one minute of dataset length RMVPE, 250 epoch, 2750 steps. Recommended to infer with RMVPE method

Kazakh News Journalist (tongue-twisting rapping voice)

Kazakh News Journalist (tongue-twisting rapping voice) (RVC v2) 250 Epoch RMVPE

Trained by me with a dataset length of almost 20 minutes. Pitch extraction used is mangio-crepe. Recommended to infer it on RMVPE or any crepe method Remember to give me credit when using this model leejhay's digital voicebox on youtube leejhay_stanpop on tiktok

Engelbert Humperdinck

Engelbert Humperdinck (RVC v2) 475 Epoch

One of the most famous singer that heard on Tiktok PH. Trained by me with a dataset length of around 23 minutes. Pitch extraction used is mangio-crepe with a hop length of 16. Huggingface link (250 epoch): Huggingface link (350 epoch): For those who want to use this model, please don't forget to give credits to: leejhay's digital voicebox on youtube *but be aware! recommended to not upload any copyrighted instrumental songs and its ai voice there as they will copyright strike your account. test it on your device locally* leejhay_stanpop on tiktok *same warning above*

Zack Tabudlo 250

Zack Tabudlo (Filipino Singer) (RVC v2) 250 & 350 Epoch

Trained by me using dataset length of 3 minutes from album including some from the latest comeback. f0 is mangio-crepe with hop length of 24 and has 3045 steps. Voice quality sounds better when inferring it on crepe and the latest rmvpe but may encounter some robotic breathing depending on your reference vocal quality. Huggingface those who want to use this model, please don't forget to give credits to: leejhay's digital voicebox on youtube *but be aware! recommended to not upload any copyrighted instrumental songs and its ai voice there as they will copyright strike your account. test it on your device locally* leejhay_stanpop on tiktok *same warning above*

Kang Haerin (from NewJeans)

Kang Haerin (from NewJeans) (RVC v2) 203 Epoch

Both of these were trained by me using AI-enhanced dataset to remove background noises and get studio quality as good as possible. May not be accurate to the original voices heard on movies but still sounds alike to those, but without echoes and reverbs. Marty McFly (Michael Fox) 360 epoch, hop length 64 Doc Emmett Brown (Christopher Lloyd) 460 epoch, hop length 32 No needed to give me credits for using those models. Feel free to use. *but be aware! recommended to not upload any copyrighted instrumental songs and its ai voice there as they will copyright strike your account. test it on your device locally*

Danielle (from NewJeans)

Description

Samples

Saiba Momoi (Blue Archive)

Adel imam (Egyptian Actor) عادل امام (ممثل مصري)

JP Furina / Focalors CV: Inori Minase (Genshin Impact)

Ariana Grande AI

Hu Tao JP (Genshin Impact)

JENNIE of BLACKPINK [Strong Ver.]

Saiba Momoi (Blue Archive) (VA: Tokui Sora)

Hatsune Miku

SpongeBob SquarePants (Talking And Singing)

Takanashi Hoshino (from Blue Archive)

Satoru Gojo (JJK) [VA Yuichi Nakamura]

ENHYPEN Heeseung

Sunaokami Shiroko (Blue Archive)

Goku (Mario Castañeda) (Latin American Spanish) (UPDATE)

Villager (Minecraft)

Mortis [Brawl stars]

Jungkook (BTS)

Melodie - (Brawl Stars) - (English & Korean Voices)

Tendou Arisu (Blue Archive)

Kanye West