Repository navigation
How to fix the noise during inference time? #23
Description
Hi Jaehyeon,
May I ask how to fix the stochastic noise during inference time? I want some generated audio to be reproducable, thus need to fix the random noise part.
Currently it seems I can only control the noise scale.
sid = torch.LongTensor([1]) # speaker identity
stn_tst = get_text("Tell me the answer please", hps_ms)
with torch.no_grad():
x_tst = stn_tst.unsqueeze(0)
x_tst_lengths = torch.LongTensor([stn_tst.size(0)])
audio = net_g_ms.infer(x_tst, x_tst_lengths, sid = sid, noise_scale=1, noise_scale_w=2, length_scale=1)[0][0,0].data.float().numpy()
ipd.display(ipd.Audio(audio, rate=hps_ms.data.sampling_rate))
Activity
BridgetteSong commentedon Jul 27, 2021
always using same input in mode.py
inferfunction:z_p = m_p + torch.randn_like(m_p) * torch.exp(logs_p) * noise_scaleyou can replace
torch.randn_likewith same fixed input:xinghua-qu commentedon Jul 27, 2021
Thanks for the reply.
But it seems the dimension of
m_pvaries every time. In this case, it's impossible to replacetorch.randn_like(m_p)as a constant tensor.You can see the dimension of
m_pthat I print out several times.To my understanding, setting the hyperparameter
noise_scale_wto be zero can enable the dimension ofm_pto be a constant value. But ifnoise_scale_wis not equal to0, is there any way to reproduce the same generated audio?BridgetteSong commentedon Jul 27, 2021
Because you use a
StochasticDurationPredictorfor duration prediction and it also contains a sample process like thise_q = torch.randn(w.size(0), 2, w.size(2)).to(device=x.device, dtype=x.dtype) * x_maskyou should also fixed this like above.
CookiePPP commentedon Jul 27, 2021
Just set the seed in torch when you input your data. Use branch rng if you want the other components to be unaffected by set seed.
It will always give you the same output from the same input.
CookiePPP commentedon Jul 27, 2021
with torch.random.fork_rn:https://pytorch.org/docs/stable/random.html
And
torch.manual_seed(0)https://pytorch.org/docs/stable/notes/randomness.html
When you call the model forward should be enough for reproduction.