soroushmehr / sampleRNN_ICLR2017

SampleRNN: An Unconditional End-to-End Neural Audio Generation Model

Home Page:https://arxiv.org/abs/1612.07837

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

add table with minimum GPU / VRAM requirements, compute times and performance averages

ballerburg9005 opened this issue · comments

Crude example (not based on actual data):

rate min. batch size min. VRAM dataset training time GPU
16Khz 64 10GB 20 h 3 days RTX 3090
32Khz 64 50GB 100h 14 days A100