A 1-minute and 52-second dataset ripped directly from THE AMAZING DIGITAL CIRCUS - Ep 3: The Mystery of Mildenhall Manor on YouTube and cleaned.
Batch size: 4
Hop Length: 64
Original Pretrain
Sampled at 32k
1 messages · Page 1 of 1 (latest)
A 1-minute and 52-second dataset ripped directly from THE AMAZING DIGITAL CIRCUS - Ep 3: The Mystery of Mildenhall Manor on YouTube and cleaned.
Batch size: 4
Hop Length: 64
Original Pretrain
Sampled at 32k
To rate this model, either use /rate or click the buttons below.
I would have posted it yesterday if it weren't for my stupid job
work getting in the way of cooler work 
good model better than mine lol
in the sense just to understand do I need to lower the bacht size for voices like this?
@zinc nest can u pls send the "william" audio here or in dm's pls?
not for specific voices exactly, batch size depends more on the dataset.
for example, 1 min of data = 2 batch size, 2-4 minutes of data = 4 batch size, 6 minutes of data = 6 batch size, 10 or more minutes of data = 8 batch size
This model has been synced with Weights and is ready to use for free!