Name: DanTDM(Retrain - Cleaned)(Original)(32k) | AI RVC Model
Brand: Weights
Rating: 4.4 (1 reviews)

WebKit

Introducing the DanTDM RVC Model by Weights, an advanced AI Voice Model optimized for incredible audio fidelity. Retrained meticulously over 1,300 epochs and 11,700 steps, this model excels in producing high-quality AI music and text-to-speech applications. Utilizing state-of-the-art techniques such as RX 10 De Crackle, De Plosive, De Ess, and De Noise, coupled with precise normalization, the DanTDM RVC V2 is perfect for creating lifelike AI covers. This model, although derived from a talking voice dataset, provides unparalleled clarity and naturalness in voice conversion tasks. Elevate your AI music projects or generate captivating text-to-speech content today with our free AI tools. Let's turn your creative ideas into reality at Weights!

This is a retrain of my DanTDM 600 epoch model Trained & Tested on Applio(Kaggle) Talking model cannot handle singing voices Used RX 10 De Crackle De Plosive De Ess Voice De Noise Normalize -3.00 Minutes 56 Seconds

DanTDM(Retrain - Cleaned)(Original)(32k)

DanTDM(Retrain - Cleaned)(RVC V2)(Original)(RMVPE)(32k)(1300 Epochs - 11,700 Steps)

.yormother

Note: i was kinda bored and decided to retrain my voice again cuz the other ones of me was doo doo soo yeh. Used:UVR5-HF - RX10 - Audacity(Too lazy to type) Minutes 24 Seconds Batch Size:10 Mic Used: offbrand mic of shure

Balbal(Retrain Again :3)(HIFI-GAN)

Balbal(Retrain Again :3)(RVC)(HIFI-GAN)(RMVPE)(300 Epochs - 13200 Steps)(40k)

Note:Grabbed the dataset using Source2Viewer Used MSST:(Mel Roformer - Denoise & Kar by aufr33) RX-10:(De Crackle - De Ess - De Click & Mouth - Dialogue Deverb - Voice De Noise - De verb) Audacity:(Renegate - Noise Gate - Truncate) Minutes 36 Seconds Batch Size:8 Trained & Tested On Applio(Lightning AI) VA:Nate Bargatze

Grady(Personality Core - Aperture Desk Job)(HIFI-GAN)

Grady(Personality Core - Aperture Desk Job)(RVC)(HIFI-GAN)(RMVPE)(170 Epochs - 17850 Steps)(40k)

Note:from tommyinnit and jack's podcast "Shut Up Im Talking" Used UVR5-HF:(Melband Roformer - Kar by aufr33 and viperx - Kar by Becruily - Denoise Aggressive by aufr33) RX-10:(De Crackle - De Ess - De Click & Mouth - Spectral De Noise - Dialogue De verb) Audacity:(Noise Gate - Truncate) :00 Batch Size: 8 Trained & Tested On Applio(Colab)

Sarah Simons(Tommyinnit's Mother)(UKA)(32k)

Sarah Simons(Tommyinnit's Mother)(RVC)(UKA)(RMVPE)(350 Epochs - 8400 Steps)(32k)

Note:the dataset was from Dawko's interview and scott was on his phone. so its gonna be like the phone fnaf 1. Used: UVR5 HF:(Melband Roformer - Vocals Fullness - Kar by becruily - Kar by Aufr33-Viperx - Denoise & Bleed - Deverb & Echo Fused - FT 2 Bleedless by unwa) RX-10:(De Crackle - De Ess - De Click & Mouth - Dialogue Deverb - Voice De Noise) Audacity:(Noise Gate - Truncate - Renegate - Exported to Flac Mono 24 Bit) Min 39 Sec Batch Size: 4

Scott Cawthon(FNAF Creator)(HIFI-GAN)(32k)

Scott Cawthon(FNAF Creator)(RVC)(HIFI-GAN)(RMVPE)(820 Epochs -40180 Steps)(32k)

Note:Its a filipino song from their tv series Incognito its noisy cuz i didn't remove reverb :p Used Only UVR5 HF:(MelBand Roformer - Instrumental Fullness V3) MVSEP(Mel + SCnet XL) Batch Size:8 Minutes 41 Seconds

DanTDM(Retrain - Cleaned)(Original)(32k)

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