LauraTTS模型的训练花了多长时间？ about funcodec HOT 5 OPEN

The model is trained about 1.5 day on LibriTTS clean subset with an A800 GPU, and the batch size is 10240 tokens.

from funcodec.

on 8 A100 GPUs? The total duration is 6000 hours ?

from funcodec.

Only one A800 GPU. I think the duration of LibriTTS clean subset is about 244 hours

from funcodec.

The model is trained about 1.5 day on LibriTTS clean subset with an A800 GPU, and the batch size is 10240 tokens.

Hi @ZhihaoDU
10240 tokens？How do you calculate it?
speech_max_length / encoder_hop_length * batch_size = 40960 / 320 * 16 = 2048?

from funcodec.

@ZhihaoDU please
数据集132000个文件，如果设置batch_size=8, 两者相除等于16250，这是不是意味着如果还是保持num_iters_per_epoch=10000，这是不合适的。你可以分享一下，batch_size、num_iters_per_epoch、num_works怎么计算才比较合理？
另外你设置的input_size=1，也参与计算吗？

from funcodec.