Граф коммитов

215 Коммитов

Автор SHA1 Сообщение Дата
Eren Golge ec579d02a1 bug fix argparser 2019-10-31 15:13:39 +01:00
Eren Golge 60b6ec18fe bug fix for synthesis.py 2019-10-29 17:38:59 +01:00
Eren Golge 002991ca15 bug fixes, linter update and test updates 2019-10-29 14:28:49 +01:00
Eren Golge 89ef71ead8 bug fix tacotron2, decoder return order fixed 2019-10-29 13:32:20 +01:00
Eren Golge 5a56a2c096 bug fixes for forward backward training and load_data for parsing data_loader 2019-10-29 02:58:42 +01:00
Eren Golge e83a4b07d2 commention model outputs for tacotron, align outputs shapes of tacotron and tracotron2, merge bidirectional decoder 2019-10-28 14:51:19 +01:00
Eren Golge 9d5a5b0764 linter 2019-10-24 14:34:31 +02:00
Eren Golge ea32f2368d linter fix 2019-10-24 14:11:07 +02:00
Eren Golge 77f5fd0584 compute and config update with new attention entropy loss 2019-10-15 15:05:42 +02:00
Eren Golge 53ec066733 replace zeros() with a better alternative 2019-10-12 18:34:12 +02:00
Eren Golge 98af061d2e formatting, merge GST model with Tacotron 2019-09-24 16:18:48 +02:00
Eren Golge e8d29613f1 fix stop condition 2019-09-24 15:38:28 +02:00
Eren Golge 9a2bd7f9af fix for 2 dim memory tensor 2019-09-23 10:25:51 +02:00
Eren Golge e085c4757d bug fix 2019-09-23 10:25:51 +02:00
Eren Golge 8d3775a7d6 Update tacotron2 for gradual training and chnage the indexing of prenet nputs to oick the last frame 2019-09-23 10:25:51 +02:00
Eren Golge a1322530df integrade concatinative speker embedding to tacotron 2019-09-12 10:39:15 +02:00
Reuben Morais 3c5aeb5e22 Fix installation by using an explicit symlink 2019-08-29 11:49:53 +02:00
Eren Golge 72ad58d893 change the bitwise for masking and small fixes 2019-08-19 16:24:28 +02:00
Eren Golge c637aa04a2 pylint 2019-08-16 15:49:12 +02:00
Eren Golge 5629292bde bug fixes 2019-08-16 15:08:04 +02:00
Eren Golge 8fde0ac00e pylint 2019-08-16 14:23:26 +02:00
Eren Golge b22c7d4a29 Merge branch 'dev-gradual-queue' into dev 2019-08-16 13:20:17 +02:00
Eren Golge 64f2b95c31 update regarding torch 1.2 2019-08-13 12:14:34 +02:00
Thomas Werkmeister ab42396fbf undo loc attn after fwd attn 2019-07-25 13:04:41 +02:00
Thomas Werkmeister f3dac0aa84 updating location attn after calculating fwd attention 2019-07-24 11:49:07 +02:00
Thomas Werkmeister 40f56f9b00 simplified code for fwd attn 2019-07-24 11:47:06 +02:00
Thomas Werkmeister a6118564d5 renamed query_rnn back to attention_rnn 2019-07-24 11:46:34 +02:00
Thomas Werkmeister fb7c5b1996 unused instance vars 2019-07-23 20:02:31 +02:00
Thomas Werkmeister 82db35530f unused var 2019-07-23 19:33:56 +02:00
Thomas Werkmeister 98edb7a4f8 renamed attention_rnn to query_rnn 2019-07-23 18:38:09 +02:00
Eren Golge 2bbb3f7a40 don't use sigmoid output for tacotron, fix bug for memory queue handling, remove maxout 2019-07-22 15:09:05 +02:00
Eren Golge f038b1aa3f new way of handling memory queue and enable/disable queuing in right/wrong conditions 2019-07-22 02:10:21 +02:00
Eren Golge d47ba5d310 gradual traning with memory queue 2019-07-20 12:33:21 +02:00
Reuben Morais 11e7895329 Fix Pylint issues 2019-07-19 09:08:51 +02:00
Eren Golge 0f0ec679ec small refactoring 2019-07-16 21:15:24 +02:00
Eren Golge 72c5062c02 use memory queue if r is smaller than queue size 2019-07-16 15:18:47 +02:00
Eren Golge f8195834ee TODO added 2019-07-16 15:18:00 +02:00
Eren Golge c72470bcfc update forward attention 2019-06-24 16:57:29 +02:00
Eren Golge 88575edd5a gst_layers 2019-06-05 18:34:48 +02:00
Eren Golge d7e0f828cf remove print 2019-06-04 00:40:03 +02:00
Eren Golge 4678c66599 forward_attn_mask and config update 2019-06-04 00:39:29 +02:00
Eren Golge f774db0241 bug fix #207 2019-05-29 00:37:41 +02:00
Eren Golge 013ec2f168 bug fix for tacotron adapting it to new common layers 2019-05-28 14:29:17 +02:00
Eren Golge 0b5a00d29e enforce monotonic attention in forward attention y for batches 2019-05-28 14:28:32 +02:00
Eren Golge 35b76556e4 Use Attention and Prenet from common file 2019-05-27 15:30:57 +02:00
Eren Golge 59ba37904d enforce monotonic attention for forward attention in eval time 2019-05-27 14:41:30 +02:00
Eren Golge ba492f43be Set tacotron model parameters to adap to common_layers.py - Prenet and Attention 2019-05-27 14:40:28 +02:00
Eren Golge 2586be7d33 batch-wise operation 2019-05-24 13:40:56 +02:00
Eren Golge 3a4a3e571a Force alignment of forward attention 2019-05-24 13:18:18 +02:00
Eren Golge e62659da94 update separate stopnet flow to make it faster. 2019-05-17 16:15:43 +02:00