bbc50faef2
add generated api_doc
2020-12-18 10:54:50 +08:00
796e0b1e1f
1. add interfaces for inference;;
...
2. add a function to recursively remove weight norm;
3. wavenet: fix weight norm dimension: explicitly specify dim=1 instead of -1.
2020-12-12 18:21:20 +08:00
b12eda8423
add network of tacotron2 model
2020-12-09 09:08:17 +00:00
62959759f9
add linear in decoder prenet
2020-12-05 22:09:44 +08:00
d3761683e1
add an adaptive loss to balance stop prediction classes
2020-12-05 14:12:30 +08:00
a4a0bd8c98
add last bn for the decoder postnet, switch back to weighted mean
2020-12-05 14:00:08 +08:00
c57e8e7350
fix transformer_tts' stop condition
2020-12-04 02:11:02 +08:00
4df5ad42f6
remove the last layer from decoder prenet
2020-12-03 15:55:07 +08:00
810f979dba
siwtch to keras style sample_weight in losses
2020-12-03 15:37:43 +08:00
9cb5c03069
transformer_tts, miscellaneous fixes
2020-12-01 18:13:30 +08:00
0cdad602e2
fix a bug for changing reduction factor in transformner_tts
2020-11-03 11:18:46 +08:00
45d6f3b99d
specify a U(-.05, .05) initializer for Embedding
2020-10-30 17:42:06 +08:00
57d820f055
add support for channel last in batch_spec, and Conv1dBatchNorm
2020-10-30 15:13:57 +08:00
36cc543348
minor fixes to TransformerTTS
2020-10-28 11:05:47 +08:00
c43216ae9b
1. API renaming Conv1d -> Conv1D, BatchNorm1d -> BatchNorm1D;
...
2. add losses in parakeet/modules;
3. fix a bug in phonetics;
4. TransformerTTS update: encoder dim can be different from decoder dim;
5. MultiHeadAttention in TransformerTTS: add k_input_dim & v_input_dim in __init__ to allow differemt feature sizes for k and v.
2020-10-22 05:04:45 +00:00
c1e0aecdde
1. import models into parakeet.models;
...
2. add predict for TransformerTTS and test its io.
2020-10-16 13:51:56 +08:00
5270774bb0
tested io for TransformerTTS
2020-10-15 22:48:09 +08:00
40457227e6
move Conv1dBatchNorm to conv.py
2020-10-14 10:05:26 +08:00
f9087ea9a2
add masking functions
2020-10-13 15:53:18 +08:00
a8192c79cc
WIP: refactor
2020-10-10 15:51:54 +08:00
iclementine
lfchener