eden-emu/opus
15
0
Fork 0
mirror of https://github.com/xiph/opus.git synced 2025-05-29 06:39:15 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

stashing stuff here

Browse source
This commit is contained in:
Jean-Marc Valin 2018-07-12 18:20:25 -04:00
parent 679dfbab58
commit 374ba430c4
4 changed files with 21 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

1
dnn/denoise.c
View file

@ -577,6 +577,7 @@ int main(int argc, char **argv) {
return 0; return 0;
} }
for (i=0;i<FRAME_SIZE;i++) x[i] = tmp[i]; for (i=0;i<FRAME_SIZE;i++) x[i] = tmp[i];
for (i=0;i<FRAME_SIZE;i++) x[i] += rand()/(float)RAND_MAX - .5;
for (i=0;i<FRAME_SIZE;i++) E += tmp[i]*(float)tmp[i]; for (i=0;i<FRAME_SIZE;i++) E += tmp[i]*(float)tmp[i];
biquad(x, mem_hp_x, x, b_hp, a_hp, FRAME_SIZE); biquad(x, mem_hp_x, x, b_hp, a_hp, FRAME_SIZE);
preemphasis(x, &mem_preemph, x, PREEMPHASIS, FRAME_SIZE); preemphasis(x, &mem_preemph, x, PREEMPHASIS, FRAME_SIZE);

20
dnn/lpcnet.py
View file

@ -12,7 +12,7 @@ import sys
rnn_units=512 rnn_units=512
pcm_bits = 8 pcm_bits = 8
pcm_levels = 2**pcm_bits pcm_levels = 2**pcm_bits
nb_used_features = 37 nb_used_features = 38
def new_wavernn_model(): def new_wavernn_model():
@ -22,11 +22,11 @@ def new_wavernn_model():
dec_feat = Input(shape=(None, 32)) dec_feat = Input(shape=(None, 32))
dec_state = Input(shape=(rnn_units,)) dec_state = Input(shape=(rnn_units,))
conv1 = Conv1D(16, 7, padding='causal') conv1 = Conv1D(16, 7, padding='causal', activation='tanh')
pconv1 = Conv1D(16, 5, padding='same') pconv1 = Conv1D(16, 5, padding='same', activation='tanh')
pconv2 = Conv1D(16, 5, padding='same') pconv2 = Conv1D(16, 5, padding='same', activation='tanh')
fconv1 = Conv1D(128, 3, padding='same') fconv1 = Conv1D(128, 3, padding='same', activation='tanh')
fconv2 = Conv1D(32, 3, padding='same') fconv2 = Conv1D(32, 3, padding='same', activation='tanh')
if False: if False:
cpcm = conv1(pcm) cpcm = conv1(pcm)
@ -40,17 +40,17 @@ def new_wavernn_model():
rep = Lambda(lambda x: K.repeat_elements(x, 160, 1)) rep = Lambda(lambda x: K.repeat_elements(x, 160, 1))
rnn = CuDNNGRU(rnn_units, return_sequences=True, return_state=True) rnn = CuDNNGRU(rnn_units, return_sequences=True, return_state=True)
rnn_in = Concatenate()([cpcm, cpitch, rep(cfeat)]) rnn_in = Concatenate()([cpcm, rep(cfeat)])
md = MDense(pcm_levels, activation='softmax') md = MDense(pcm_levels, activation='softmax')
gru_out, state = rnn(rnn_in) gru_out, state = rnn(rnn_in)
ulaw_prob = md(gru_out) ulaw_prob = md(gru_out)
model = Model([pcm, pitch, feat], ulaw_prob) model = Model([pcm, feat], ulaw_prob)
encoder = Model(feat, cfeat) encoder = Model(feat, cfeat)
dec_rnn_in = Concatenate()([cpcm, cpitch, dec_feat]) dec_rnn_in = Concatenate()([cpcm, dec_feat])
dec_gru_out, state = rnn(dec_rnn_in, initial_state=dec_state) dec_gru_out, state = rnn(dec_rnn_in, initial_state=dec_state)
dec_ulaw_prob = md(dec_gru_out) dec_ulaw_prob = md(dec_gru_out)
decoder = Model([pcm, pitch, dec_feat, dec_state], [dec_ulaw_prob, state]) decoder = Model([pcm, dec_feat, dec_state], [dec_ulaw_prob, state])
return model, encoder, decoder return model, encoder, decoder

9
dnn/test_lpcnet.py
View file

@ -47,7 +47,7 @@ in_data = np.reshape(in_data, (nb_frames*pcm_chunk_size, 1))
out_data = np.reshape(data, (nb_frames*pcm_chunk_size, 1)) out_data = np.reshape(data, (nb_frames*pcm_chunk_size, 1))
model.load_weights('lpcnet1i_30.h5') model.load_weights('lpcnet3a_21.h5')
order = 16 order = 16
@ -61,7 +61,7 @@ for c in range(1, nb_frames):
cfeat = enc.predict(features[c:c+1, :, :nb_used_features]) cfeat = enc.predict(features[c:c+1, :, :nb_used_features])
for fr in range(1, feature_chunk_size): for fr in range(1, feature_chunk_size):
f = c*feature_chunk_size + fr f = c*feature_chunk_size + fr
a = features[c, fr, nb_used_features+1:] a = features[c, fr, nb_used_features:]
#print(a) #print(a)
gain = 1.; gain = 1.;
@ -69,9 +69,10 @@ for c in range(1, nb_frames):
period = period - 4 period = period - 4
for i in range(frame_size): for i in range(frame_size):
pitch[0, 0, 0] = exc[f*frame_size + i - period, 0] pitch[0, 0, 0] = exc[f*frame_size + i - period, 0]
fexc[0, 0, 0] = exc[f*frame_size + i - 1] fexc[0, 0, 0] = 2*exc[f*frame_size + i - 1]
#fexc[0, 0, 0] = in_data[f*frame_size + i, 0]
#print(cfeat.shape) #print(cfeat.shape)
p, state = dec.predict([fexc, pitch, cfeat[:, fr:fr+1, :], state]) p, state = dec.predict([fexc, cfeat[:, fr:fr+1, :], state])
p = p/(1e-5 + np.sum(p)) p = p/(1e-5 + np.sum(p))
#print(np.sum(p)) #print(np.sum(p))
iexc[0, 0, 0] = np.argmax(np.random.multinomial(1, p[0,0,:], 1))-128 iexc[0, 0, 0] = np.argmax(np.random.multinomial(1, p[0,0,:], 1))-128

10
dnn/train_lpcnet.py
View file

@ -13,13 +13,13 @@ from adadiff import Adadiff
import tensorflow as tf import tensorflow as tf
from keras.backend.tensorflow_backend import set_session from keras.backend.tensorflow_backend import set_session
config = tf.ConfigProto() config = tf.ConfigProto()
config.gpu_options.per_process_gpu_memory_fraction = 0.28 config.gpu_options.per_process_gpu_memory_fraction = 0.44
set_session(tf.Session(config=config)) set_session(tf.Session(config=config))
nb_epochs = 40 nb_epochs = 40
batch_size = 64 batch_size = 64
model = lpcnet.new_wavernn_model() model, enc, dec = lpcnet.new_wavernn_model()
model.compile(optimizer=Adadiff(), loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy']) model.compile(optimizer=Adadiff(), loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy'])
model.summary() model.summary()
@ -63,8 +63,8 @@ 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('lpcnet1k_{epoch:02d}.h5') checkpoint = ModelCheckpoint('lpcnet3b_{epoch:02d}.h5')
#model.load_weights('wavernn1c_01.h5') #model.load_weights('wavernn1c_01.h5')
model.compile(optimizer=Adadiff(), 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'])
model.fit([in_data, in_pitch, features], out_data, batch_size=batch_size, epochs=30, validation_split=0.2, callbacks=[checkpoint]) model.fit([in_data, features], out_data, batch_size=batch_size, epochs=30, validation_split=0.2, callbacks=[checkpoint])