Keep proj_dim in speaker encoder models

This commit is contained in:
Eren Gölge 2022-01-03 15:03:34 +00:00
parent 03bcae1ba5
commit 5840d89802
2 changed files with 2 additions and 0 deletions

View File

@ -49,6 +49,7 @@ class LSTMSpeakerEncoder(nn.Module):
self.use_lstm_with_projection = use_lstm_with_projection
self.use_torch_spec = use_torch_spec
self.audio_config = audio_config
self.proj_dim = proj_dim
layers = []
# choise LSTM layer

View File

@ -95,6 +95,7 @@ class ResNetSpeakerEncoder(nn.Module):
self.log_input = log_input
self.use_torch_spec = use_torch_spec
self.audio_config = audio_config
self.proj_dim = proj_dim
self.conv1 = nn.Conv2d(1, num_filters[0], kernel_size=3, stride=1, padding=1)
self.relu = nn.ReLU(inplace=True)