Malar Kannan
|
f44665e9b2
|
1. fixed softmax output and overfit the model for small sample
2. updated to run on complete data
|
2017-12-12 12:18:27 +05:30 |
|
Malar Kannan
|
cc4fbe45b9
|
trying to overfit 2 samples with model -> doesn't seem to converge
|
2017-12-11 15:03:14 +05:30 |
|
Malar Kannan
|
8d550c58cc
|
fixed batch normalization layer before activation
|
2017-12-11 14:33:56 +05:30 |
|
Malar Kannan
|
240ecb3f27
|
removed bn output layer
|
2017-12-11 14:12:23 +05:30 |
|
Malar Kannan
|
05242d5991
|
added batch normalization
|
2017-12-11 14:09:04 +05:30 |
|
Malar Kannan
|
fea9184aec
|
using the full data and fixed typo in model layer name
|
2017-12-11 13:47:30 +05:30 |
|
Malar Kannan
|
a6543491f8
|
fixed empty phoneme boundary case
|
2017-12-11 13:05:46 +05:30 |
|
Malar Kannan
|
d387922f7d
|
added dense-relu/softmax layers to segment output
|
2017-12-11 12:30:08 +05:30 |
|
Malar Kannan
|
52bbb69c65
|
resuming segment training
|
2017-12-10 21:58:55 +05:30 |
|
Malar Kannan
|
03edd935ea
|
fixed input_dim
|
2017-12-07 17:16:05 +05:30 |
|
Malar Kannan
|
a7f1451a7f
|
fixed exception in data generation
|
2017-12-07 16:49:34 +05:30 |
|
Malar Kannan
|
91fde710f3
|
completed the segmentation model
|
2017-12-07 15:17:59 +05:30 |
|
Malar Kannan
|
c8a07b3d7b
|
Merge branch 'master' of ssh://invnuc/~/Public/Repos/speech_scoring
|
2017-12-07 12:00:59 +05:30 |
|
Malar Kannan
|
8785522196
|
Merge branch 'master' of /home/ilml/Public/Repos/speech_scoring
|
2017-12-07 12:00:44 +05:30 |
|
Malar Kannan
|
435c4a4aa6
|
added a resume parameter for training
|
2017-12-07 12:00:42 +05:30 |
|
Malar Kannan
|
c1801b5aa3
|
implented segment tfrecords batch data-generator
|
2017-12-07 11:48:19 +05:30 |
|
Malar Kannan
|
c0369d7a66
|
Merge branch 'master' of ssh://gpuaws/~/repos/speech_scoring
|
2017-12-06 17:33:27 +05:30 |
|
Malar Kannan
|
8e14db2437
|
Merge branch 'master' of ssh://invmac/~/Public/repos/speech-scoring
|
2017-12-06 17:32:46 +05:30 |
|
Malar Kannan
|
bcf1041bde
|
created segment sample tfrecord writer
|
2017-12-06 17:32:26 +05:30 |
|
Malar Kannan
|
b50edb980d
|
implemented segment-generation for random words for testing
|
2017-12-06 14:41:25 +05:30 |
|
Malar Kannan
|
3f76207f0d
|
using pitch contour instead of spectrogram
|
2017-12-04 19:15:17 +05:30 |
|
Malar Kannan
|
6ef4e86f41
|
implemented segmentation visualization
|
2017-11-30 14:49:55 +05:30 |
|
Malar Kannan
|
0b1152b5c3
|
implemented the model, todo implement ctc and training queueing logic
|
2017-11-28 19:10:19 +05:30 |
|
Malar Kannan
|
1928fce4e8
|
Merge branch 'master' of ssh://invnuc/~/Public/Repos/speech_scoring
|
2017-11-28 17:05:35 +05:30 |
|
Malar Kannan
|
ec7303223c
|
merged
|
2017-11-28 17:05:20 +05:30 |
|
Malar Kannan
|
f12da988d3
|
segmentation model wip
|
2017-11-28 15:46:39 +05:30 |
|
Malar Kannan
|
705cf3d172
|
finding exact duration of sound sample
|
2017-11-28 12:52:20 +05:30 |
|
Malar Kannan
|
8f79316893
|
Merge branch 'master' of /Users/malarkannan/Public/repos/speech-scoring
|
2017-11-28 12:32:50 +05:30 |
|
Malar Kannan
|
0345cc46ae
|
implemented tts sementation generation code
|
2017-11-28 12:32:45 +05:30 |
|
Malar Kannan
|
20b2d7a958
|
updated model data
|
2017-11-27 14:08:01 +05:30 |
|
Malar Kannan
|
43d5b75db9
|
removing spec_n counter
|
2017-11-24 11:06:42 +00:00 |
|
Malar Kannan
|
ec08cc7d62
|
Merge branch 'master' of ssh://gpuaws/~/repos/speech_scoring
|
2017-11-24 14:32:43 +05:30 |
|
Malar Kannan
|
2268ad8bb0
|
implemented pitch plotting
|
2017-11-24 14:32:13 +05:30 |
|
Malar Kannan
|
ec317b6628
|
Merge branch 'master' of /home/ilml/Public/Repos/speech_scoring
|
2017-11-24 14:26:40 +05:30 |
|
Malar Kannan
|
235300691e
|
find spec_n from tfrecords
|
2017-11-24 14:26:36 +05:30 |
|
Malar Kannan
|
ae46578aec
|
Merge branch 'master' of ssh://invmac/~/Public/repos/speech-scoring
|
2017-11-23 17:50:47 +05:30 |
|
Malar Kannan
|
3d7542271d
|
implemented tts segmentation data generation
|
2017-11-23 17:50:11 +05:30 |
|
Malar Kannan
|
54f38ca775
|
removed a layer using lstm
|
2017-11-22 15:46:42 +05:30 |
|
Malar Kannan
|
6355db4af7
|
adding missing model-dir for training constants copying
|
2017-11-22 15:04:02 +05:30 |
|
Malar Kannan
|
1f60183ab8
|
Merge branch 'master' of ssh://invnuc/~/Public/Repos/speech_scoring
|
2017-11-22 14:45:35 +05:30 |
|
Malar Kannan
|
e7fc607578
|
trying mfcc instead of spectrogram
|
2017-11-22 14:45:08 +05:30 |
|
Malar Kannan
|
d2a075422c
|
copying constantws to models
|
2017-11-20 15:15:27 +05:30 |
|
Malar Kannan
|
a5d4ede35d
|
finding number of record by streaming-onepass
|
2017-11-20 12:07:13 +05:30 |
|
Malar Kannan
|
3ae8dc50a2
|
implemented pair data inspection
|
2017-11-17 17:29:48 +05:30 |
Ubuntu
|
c81a7b4468
|
decreasing first layer node count to avoid gpu memory overflow
|
2017-11-17 10:31:36 +00:00 |
|
Malar Kannan
|
c682962c8f
|
using a Bi-LSTM layer as the first layer
|
2017-11-17 14:17:12 +05:30 |
|
Malar Kannan
|
6ff052be9b
|
fixed randomize pair picking
|
2017-11-17 11:57:38 +05:30 |
|
Malar Kannan
|
7fc89c0853
|
1. fixed pairing and data duplicates
2. clean-up
|
2017-11-16 23:41:38 +05:30 |
|
Malar Kannan
|
3d297f176f
|
perfect score on new test words - TODO evaluate on real voice
|
2017-11-16 14:19:25 +05:30 |
|
Malar Kannan
|
7d94ddc2ae
|
all phrases
|
2017-11-15 18:30:49 +05:30 |