shithub: opus

--- a/dnn/training_tf2/fec_encoder.py

+++ b/dnn/training_tf2/fec_encoder.py

@@ -44,7 +44,7 @@

     args = parser.parse_args()

-model, encoder, decoder = new_rdovae_model(nb_used_features=20, nb_bits=80, batch_size=1, cond_size=args.cond_size)

+model, encoder, decoder, qembedding = new_rdovae_model(nb_used_features=20, nb_bits=80, batch_size=1, cond_size=args.cond_size)

 model.load_weights(args.weights)

 lpc_order = 16

@@ -106,26 +106,30 @@

 features = features[:, :, :nb_used_features]

 features = features[:, :num_subframes, :]

-# lambda and q_id (ToDo: check validity of lambda and q_id)

-enc_lambda = args.enc_lambda * np.ones((1, num_frames, 1))

-quant_id = np.round(10*np.log(enc_lambda/.0007)).astype('int16')

+#variable quantizer depending on the delay

+q0 = 2

+q1 = 10

+quant_id = np.round(q1 + (q0-q1)*np.arange(args.num_redundancy_frames//2)/args.num_redundancy_frames).astype('int16')

+#print(quant_id)

+quant_embed = qembedding(quant_id)

 # run encoder

 print("running fec encoder...")

-symbols, quant_embed_dec, gru_state_dec = encoder.predict([features, quant_id, enc_lambda])

+symbols, gru_state_dec = encoder.predict(features)

 # apply quantization

 nsymbols = 80

-dead_zone = tf.math.softplus(quant_embed_dec[:, :, nsymbols : 2 * nsymbols])

-symbols = apply_dead_zone([symbols, dead_zone]).numpy()

-qsymbols = np.round(symbols)

-quant_gru_state_dec = pvq_quantize(gru_state_dec, 30)

+quant_scale = tf.math.softplus(quant_embed[:, :nsymbols]).numpy()

+dead_zone = tf.math.softplus(quant_embed[:, nsymbols : 2 * nsymbols]).numpy()

+#symbols = apply_dead_zone([symbols, dead_zone]).numpy()

+#qsymbols = np.round(symbols)

+quant_gru_state_dec = pvq_quantize(gru_state_dec, 82)

 # rate estimate

-hard_distr_embed = tf.math.sigmoid(quant_embed_dec[:, :, 4 * nsymbols : ]).numpy()

-rate_input = np.concatenate((qsymbols, hard_distr_embed, enc_lambda), axis=-1)

-rates = sq_rate_metric(None, rate_input, reduce=False).numpy()

+hard_distr_embed = tf.math.sigmoid(quant_embed[:, 4 * nsymbols : ]).numpy()

+#rate_input = np.concatenate((qsymbols, hard_distr_embed, enc_lambda), axis=-1)

+#rates = sq_rate_metric(None, rate_input, reduce=False).numpy()

 # run decoder

 input_length = args.num_redundancy_frames // 2

@@ -134,22 +138,48 @@

 packets = []

 packet_sizes = []

+sym_batch = np.zeros((num_frames-offset, args.num_redundancy_frames//2, nsymbols), dtype='float32')

+quant_state = quant_gru_state_dec[0, offset:num_frames, :]

+#pack symbols for batch processing

 for i in range(offset, num_frames):

-    print(f"processing frame {i - offset}...")

-    features = decoder.predict([qsymbols[:, i - 2 * input_length + 2 : i + 1 : 2, :], quant_embed_dec[:, i - 2 * input_length + 2 : i + 1 : 2, :], quant_gru_state_dec[:, i, :]])

-    packets.append(features)

-    packet_size = 8 * int((np.sum(rates[:, i - 2 * input_length + 2 : i + 1 : 2]) + 7) / 8) + 64

-    packet_sizes.append(packet_size)

+    sym_batch[i-offset, :, :] = symbols[0, i - 2 * input_length + 2 : i + 1 : 2, :]

+#quantize symbols

+sym_batch = sym_batch * quant_scale

+sym_batch = apply_dead_zone([sym_batch, dead_zone]).numpy()

+sym_batch = np.round(sym_batch)

+hard_distr_embed = np.broadcast_to(hard_distr_embed, (sym_batch.shape[0], sym_batch.shape[1], 2*sym_batch.shape[2]))

+fake_lambda = np.ones((sym_batch.shape[0], sym_batch.shape[1], 1), dtype='float32')

+rate_input = np.concatenate((sym_batch, hard_distr_embed, fake_lambda), axis=-1)

+rates = sq_rate_metric(None, rate_input, reduce=False).numpy()

+print("rate = ", np.mean(rates))

+sym_batch = sym_batch / quant_scale

+print(sym_batch.shape, quant_state.shape)

+#features = decoder.predict([sym_batch, quant_state])

+features = decoder([sym_batch, quant_state])

+#for i in range(offset, num_frames):

+#    print(f"processing frame {i - offset}...")

+#    features = decoder.predict([qsymbols[:, i - 2 * input_length + 2 : i + 1 : 2, :], quant_embed_dec[:, i - 2 * input_length + 2 : i + 1 : 2, :], quant_gru_state_dec[:, i, :]])

+#    packets.append(features)

+#    packet_size = 8 * int((np.sum(rates[:, i - 2 * input_length + 2 : i + 1 : 2]) + 7) / 8) + 64

+#    packet_sizes.append(packet_size)

 # write packets

 packet_file = args.output + '.fec' if not args.output.endswith('.fec') else args.output

-write_fec_packets(packet_file, packets, packet_sizes)

+#write_fec_packets(packet_file, packets, packet_sizes)

-print(f"average redundancy rate: {int(round(sum(packet_sizes) / len(packet_sizes) * 50 / 1000))} kbps")

+#print(f"average redundancy rate: {int(round(sum(packet_sizes) / len(packet_sizes) * 50 / 1000))} kbps")

+#create packets array like in the original version for debugging purposes

+for i in range(offset, num_frames):

+    packets.append(features[i-offset:i-offset+1, :, :])

 if args.debug_output:

     import itertools

@@ -160,6 +190,7 @@

     for batch, offset in itertools.product(batches, offsets):

         stop = packets[0].shape[1] - offset

+        print(batch, offset, stop)

         test_features = np.concatenate([packet[:,stop - batch: stop, :] for packet in packets[::batch//2]], axis=1)

         test_features_full = np.zeros((test_features.shape[1], nb_features), dtype=np.float32)

--

⑨