Skip to content

Modify configs/default.yaml #23

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 13 commits into
base: master
Choose a base branch
from
Open

Modify configs/default.yaml #23

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
cff3e22
:books: Modify configs/default.yaml
rishikksh20 Sep 14, 2020
5f1d667
Update train_fastspeech.py
carankt Sep 14, 2020
872bfdb
Add waveflow vocoder too
carankt Sep 14, 2020
d23626c
remove multi valid tb plot
carankt Sep 14, 2020
dd4c5ac
detach PE and remove PE embedding
carankt Sep 14, 2020
81b4dc8
Delete train_filelist.txt
carankt Sep 14, 2020
20c86ab
Delete valid_filelist.txt
carankt Sep 14, 2020
0e93eab
Create valid_filelist.txt
carankt Sep 14, 2020
d6fe072
Add files via upload
carankt Sep 14, 2020
c9c4055
Remove the . error
carankt Sep 14, 2020
b51c85b
Update inference.py
carankt Sep 14, 2020
e0633df
Update inference.py
carankt Sep 14, 2020
b24cfd0
Removes empty lines from input para
carankt Sep 15, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions configs/default.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ train:
# optimization related
eos: False #True
opt: 'noam'
accum_grad: 4
accum_grad: 1
grad_clip: 1.0
weight_decay: 0.001
patience: 0
Expand All @@ -126,7 +126,7 @@ train:
seed: 1 # random seed number
resume: "" # the snapshot path to resume (if set empty, no effect)
use_phonemes: True
batch_size : 16
batch_size : 48
# other
melgan_vocoder : True
save_interval : 1000
Expand Down
8 changes: 4 additions & 4 deletions fastspeech.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -211,12 +211,12 @@ def _forward(
# print("d_outs:", d_outs.shape) # torch.Size([32, 121])
hs = self.length_regulator(hs, ds, ilens) # (B, Lmax, adim)
# print("After Hs:",hs.shape) #torch.Size([32, 868, 256])
e_outs = self.energy_predictor(hs, mel_masks)
e_outs = self.energy_predictor(hs.detach(), mel_masks)
# print("e_outs:", e_outs.shape) #torch.Size([32, 868])
p_outs = self.pitch_predictor(hs, mel_masks)
p_outs = self.pitch_predictor(hs.detach(), mel_masks)
# print("p_outs:", p_outs.shape) #torch.Size([32, 868])
hs = hs + self.pitch_embed(one_hot_pitch) # (B, Lmax, adim)
hs = hs + self.energy_embed(one_hot_energy) # (B, Lmax, adim)
hs = hs + one_hot_pitch + one_hot_energy # self.pitch_embed(one_hot_pitch) (B, Lmax, adim)
#hs = hs + self.energy_embed(one_hot_energy) # (B, Lmax, adim)
# forward decoder
if olens is not None:
h_masks = self._source_mask(olens)
Expand Down
Loading