ref: c0cfcbcfbd8e5625a2aa164340bab685597d5be0
parent: e7f2360936991ec6e2a96a7d29eaf21db783f4dc
author: Jean-Marc Valin <jmvalin@amazon.com>
date: Mon Dec 19 12:08:16 EST 2022
update model New model that can handle absent Burg features
--- a/dnn/autogen.sh
+++ b/dnn/autogen.sh
@@ -6,7 +6,7 @@
test -n "$srcdir" && cd "$srcdir"
#SHA1 of the first commit compatible with the current model
-commit=97e64b3
+commit=5e88ebe
./download_model.sh $commit
echo "Updating build configuration files for lpcnet, please wait...."
--- a/dnn/training_tf2/plc_loader.py
+++ b/dnn/training_tf2/plc_loader.py
@@ -47,19 +47,25 @@
def __getitem__(self, index):
features = self.features[self.indices[index*self.batch_size:(index+1)*self.batch_size], :, :]
- #lost = (np.random.rand(features.shape[0], features.shape[1]) > .2).astype('float')+ burg_lost = (np.random.rand(features.shape[0], features.shape[1]) > .1).astype('float')+ burg_lost = np.reshape(burg_lost, (features.shape[0], features.shape[1], 1))
+ burg_mask = np.tile(burg_lost, (1,1,self.nb_burg_features))
+
lost = self.lost_offset[self.lost_indices[index*self.batch_size:(index+1)*self.batch_size], :]
lost = np.reshape(lost, (features.shape[0], features.shape[1], 1))
lost_mask = np.tile(lost, (1,1,features.shape[2]))
in_features = features*lost_mask
+ in_features[:,:,:self.nb_burg_features] = in_features[:,:,:self.nb_burg_features]*burg_mask
#For the first frame after a loss, we don't have valid features, but the Burg estimate is valid.
- in_features[:,1:,self.nb_burg_features:] = in_features[:,1:,self.nb_burg_features:]*lost_mask[:,:-1,self.nb_burg_features:]
+ #in_features[:,1:,self.nb_burg_features:] = in_features[:,1:,self.nb_burg_features:]*lost_mask[:,:-1,self.nb_burg_features:]
out_lost = np.copy(lost)
- out_lost[:,1:,:] = out_lost[:,1:,:]*out_lost[:,:-1,:]
+ #out_lost[:,1:,:] = out_lost[:,1:,:]*out_lost[:,:-1,:]
out_features = np.concatenate([features[:,:,self.nb_burg_features:], 1.-out_lost], axis=-1)
- inputs = [in_features*lost_mask, lost]
+ burg_sign = 2*burg_lost - 1
+ # last dim is 1 for received packet, 0 for lost packet, and -1 when just the Burg info is missing
+ inputs = [in_features*lost_mask, lost*burg_sign]
outputs = [out_features]
return (inputs, outputs)
--
⑨