eden-emu/opus
15
0
Fork 0
mirror of https://github.com/xiph/opus.git synced 2025-05-19 09:58:30 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Add prediction

Browse source
This commit is contained in:
Jean-Marc Valin 2018-08-14 18:40:32 -04:00
parent 4fec1144f3
commit 3d20cdaed4
2 changed files with 19 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

6
dnn/lpcnet.py
View file

@ -10,7 +10,7 @@ import numpy as np
import h5py import h5py
import sys import sys
rnn_units=512 rnn_units=64
pcm_bits = 8 pcm_bits = 8
pcm_levels = 2**pcm_bits pcm_levels = 2**pcm_bits
nb_used_features = 38 nb_used_features = 38
@ -41,7 +41,7 @@ class PCMInit(Initializer):
} }
def new_wavernn_model(): def new_wavernn_model():
pcm = Input(shape=(None, 1)) pcm = Input(shape=(None, 2))
pitch = Input(shape=(None, 1)) pitch = Input(shape=(None, 1))
feat = Input(shape=(None, nb_used_features)) feat = Input(shape=(None, nb_used_features))
dec_feat = Input(shape=(None, 32)) dec_feat = Input(shape=(None, 32))
@ -61,7 +61,7 @@ def new_wavernn_model():
cpitch = pitch cpitch = pitch
embed = Embedding(256, 128, embeddings_initializer=PCMInit()) embed = Embedding(256, 128, embeddings_initializer=PCMInit())
cpcm = Reshape((-1, 128))(embed(pcm)) cpcm = Reshape((-1, 128*2))(embed(pcm))
cfeat = fconv2(fconv1(feat)) cfeat = fconv2(fconv1(feat))

20
dnn/train_wavenet_audio.py
View file

@ -25,16 +25,18 @@ model, _, _ = lpcnet.new_wavernn_model()
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy']) model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy'])
model.summary() model.summary()
pcmfile = sys.argv[1] exc_file = sys.argv[1]
feature_file = sys.argv[2] feature_file = sys.argv[2]
pred_file = sys.argv[3]
pcm_file = sys.argv[4]
frame_size = 160 frame_size = 160
nb_features = 54 nb_features = 54
nb_used_features = wavenet.nb_used_features nb_used_features = wavenet.nb_used_features
feature_chunk_size = 15 feature_chunk_size = 15
pcm_chunk_size = frame_size*feature_chunk_size pcm_chunk_size = frame_size*feature_chunk_size
data = np.fromfile(pcmfile, dtype='int16') data = np.fromfile(pcm_file, dtype='int16')
data = np.minimum(127, lin2ulaw(data[80:]/32768.)) data = np.minimum(127, lin2ulaw(data/32768.))
nb_frames = len(data)//pcm_chunk_size nb_frames = len(data)//pcm_chunk_size
features = np.fromfile(feature_file, dtype='float32') features = np.fromfile(feature_file, dtype='float32')
@ -46,6 +48,13 @@ in_data = np.concatenate([data[0:1], data[:-1]]);
in_data = in_data + np.random.randint(-1, 1, len(data)) in_data = in_data + np.random.randint(-1, 1, len(data))
features = np.reshape(features, (nb_frames*feature_chunk_size, nb_features)) features = np.reshape(features, (nb_frames*feature_chunk_size, nb_features))
pred = np.fromfile(pred_file, dtype='int16')
pred = pred[:nb_frames*pcm_chunk_size]
pred = np.minimum(127, lin2ulaw(pred/32768.))
pred = pred + np.random.randint(-1, 1, len(data))
pitch = 1.*data pitch = 1.*data
pitch[:320] = 0 pitch[:320] = 0
for i in range(2, nb_frames*feature_chunk_size): for i in range(2, nb_frames*feature_chunk_size):
@ -60,7 +69,10 @@ out_data = np.reshape(data, (nb_frames, pcm_chunk_size, 1))
out_data = (out_data.astype('int16')+128).astype('uint8') out_data = (out_data.astype('int16')+128).astype('uint8')
features = np.reshape(features, (nb_frames, feature_chunk_size, nb_features)) features = np.reshape(features, (nb_frames, feature_chunk_size, nb_features))
features = features[:, :, :nb_used_features] features = features[:, :, :nb_used_features]
pred = np.reshape(pred, (nb_frames, pcm_chunk_size, 1))
pred = (pred.astype('int16')+128).astype('uint8')
in_data = np.concatenate([in_data, pred], axis=-1)
#in_data = np.concatenate([in_data, in_pitch], axis=-1) #in_data = np.concatenate([in_data, in_pitch], axis=-1)
@ -68,7 +80,7 @@ features = features[:, :, :nb_used_features]
# f.create_dataset('data', data=in_data[:50000, :, :]) # f.create_dataset('data', data=in_data[:50000, :, :])
# f.create_dataset('feat', data=features[:50000, :, :]) # f.create_dataset('feat', data=features[:50000, :, :])
checkpoint = ModelCheckpoint('wavenet3g_{epoch:02d}.h5') checkpoint = ModelCheckpoint('wavenet3h9_{epoch:02d}.h5')
#model.load_weights('wavernn1c_01.h5') #model.load_weights('wavernn1c_01.h5')
model.compile(optimizer=Adam(0.001, amsgrad=True, decay=2e-4), loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy']) model.compile(optimizer=Adam(0.001, amsgrad=True, decay=2e-4), loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy'])