Microsoft David TTS
RVC v2TTS / Realtime

πŸ“£

From the creator

"Trained using 22 minutes of data. 48k sample rate rmvpe Please do credit me (@iam3m377 on youtube) I'd love to see what people do with my models it's 150 epochs because it overtrained when i went to 400 and the last backup i had was 150 (i closed the colab before checking)"

Description

Introducing the Microsoft David TTS (RVC V2) 150 Epochs model, powered by the advanced RVC framework. This text-to-speech model boasts exceptional clarity, having been trained on 22 minutes of high-quality audio at a 48k sample rate. The resulting model delivers natural-sounding speech, perfect for creating AI covers or generating realistic voiceovers. Despite initial overtraining during 400 epochs, we settled on the optimal 150 epoch version, providing an ideal balance between performance and quality. Give it a try today and discover how easy it is to use our cutting-edge AI technology!

See more

Weekly Metrics

Samples

New
Classic

Pitch

0

#1

Male

Singing

πŸ‡ΊπŸ‡Έ English

#2

Female

Singing

πŸ‡ΊπŸ‡Έ English

#3

Female

Singing (Dry)

πŸ‡ΊπŸ‡Έ English

#4

Female

Singing (High)

πŸ‡ΊπŸ‡Έ English

#5

Male

Singing 2

πŸ‡ΊπŸ‡Έ English

#6

Male

Singing (Dry)

πŸ‡ΊπŸ‡Έ English

#7

Male

Singing (Dry, High)

πŸ‡ΊπŸ‡Έ English

Comments

samples in 1-2 minutes

pretty sure glitchiness was just the ai replicating the glitchiness of the actual tts

japanese sample

Audio removed in response to copyright claim.

oh i thought it was a different tts nvm

i was thinking like microsoft sam or smth

english sample

Audio removed in response to copyright claim.

Why did you use the david madson design?

i forgot to add the instrumental

first result on google images

I mean if you really wanted to the samjoe404 designs at least look completely and not like bad paint drawings.

with instrumental

Audio removed in response to copyright claim.

Add a comment

Post