Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

make dropout at prenet optional

This commit is contained in:
Eren Golge 2019-05-12 17:34:57 +02:00
parent fe14947b0e
commit 820d18c922
3 changed files with 9 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.compute
View File

@ -9,6 +9,6 @@ pip3 install https://download.pytorch.org/whl/cu100/torch-1.0.1.post2-cp36-cp36m
wget https://www.dropbox.com/s/wqn5v3wkktw9lmo/install.sh?dl=0 -O install.sh
sudo sh install.sh
python3 setup.py develop
python3 distribute.py --config_path config_cluster.json --data_path ${USER_DIR}/MozillaAll/Mozilla/ --restore_path ${USER_DIR}/checkpoint_123000_4761.pth.tar
python3 distribute.py --config_path config_cluster.json --data_path ${USER_DIR}/MozillaAll2/Mozilla/ --restore_path ${USER_DIR}/checkpoint_123000_4761.pth.tar
# python3 distribute.py --config_path config_cluster.json --data_path ${SHARED_DIR}/data/mozilla/Judy/
# while true; do sleep 1000000; done

7
config_cluster.json
View File

@ -1,6 +1,6 @@
{
"run_name": "mozilla-fattn-no_loc",
"run_description": "Finetune 4761 with BN",
"run_name": "mozilla-fattn",
"run_description": "Finetune 4761 with BN + Dropout. It is to compare to 4780 and see how dropout behaves with BN.",
"audio":{
// Audio processing parameters
@ -41,6 +41,7 @@
"memory_size": 5, // ONLY TACOTRON - memory queue size used to queue network predictions to feed autoregressive connection. Useful if r < 5.
"attention_norm": "softmax", // softmax or sigmoid. Suggested to use softmax for Tacotron2 and sigmoid for Tacotron.
"prenet_type": "bn", // ONLY TACOTRON2 - "original" or "bn".
"prenet_dropout": true, // ONLY TACOTRON2 - enable/disable dropout at prenet.
"use_forward_attn": true, // ONLY TACOTRON2 - if it uses forward attention. In general, it aligns faster.
"transition_agent": false, // ONLY TACOTRON2 - enable/disable transition agent of forward attention.
"location_attn": false, // ONLY TACOTRON2 - enable_disable location sensitive attention. It is enabled for TACOTRON by default.
@ -65,7 +66,7 @@
"dataset": "mozilla", // DATASET-RELATED: one of TTS.dataset.preprocessors depending on your target dataset. Use "tts_cache" for pre-computed dataset by extract_features.py
"min_seq_len": 0, // DATASET-RELATED: minimum text length to use in training
"max_seq_len": 150, // DATASET-RELATED: maximum text length
"output_path": "/media/erogol/data_ssd/Data/models/mozilla_models/", // DATASET-RELATED: output path for all training outputs.
"output_path": "../keep/", // DATASET-RELATED: output path for all training outputs.
"num_loader_workers": 4, // number of training data loader processes. Don't set it too big. 4-8 are good values.
"num_val_loader_workers": 4, // number of evaluation data loader processes.
"phoneme_cache_path": "mozilla_us_phonemes", // phoneme computation is slow, therefore, it caches results in the given folder.

7
layers/tacotron2.py
View File

@ -53,9 +53,10 @@ class LinearBN(nn.Module):
class Prenet(nn.Module):
def __init__(self, in_features, prenet_type, out_features=[256, 256]):
def __init__(self, in_features, prenet_type, prenet_dropout, out_features=[256, 256]):
super(Prenet, self).__init__()
self.prenet_type = prenet_type
self.prenet_dropout = prenet_dropout
in_features = [in_features] + out_features[:-1]
if prenet_type == "bn":
self.layers = nn.ModuleList([
@ -70,9 +71,9 @@ class Prenet(nn.Module):
def forward(self, x):
for linear in self.layers:
if self.prenet_type == "original":
if self.prenet_dropout:
x = F.dropout(F.relu(linear(x)), p=0.5, training=self.training)
elif self.prenet_type == "bn":
else:
x = F.relu(linear(x))
return x