issues
search
srvk
/
eesen
The official repository of the Eesen project
http://arxiv.org/abs/1507.08240
Apache License 2.0
824
stars
342
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
accuracy drops
#122
madhavsund
opened
7 years ago
4
accuracy drops
#121
madhavsund
closed
7 years ago
0
ERROR compute-fbank-feats:Read()
#120
star633669
closed
7 years ago
3
Error training ctc parallel (Cuda problem)
#119
sanakhamekhem
closed
7 years ago
6
How can I do the decoding only base on the acoustic models where there is no effect of language models in EESEN ?
#118
niucheney
closed
7 years ago
3
separate the model loading and decoder
#117
madhavsund
closed
7 years ago
4
Difference in the length between labeled targets and feature samples
#116
razor1179
closed
7 years ago
1
output of train network and input of decoding lattice
#115
Sundy1219
opened
7 years ago
0
Mybranch2
#114
riebling
closed
7 years ago
0
Compressed .ark files
#113
razor1179
closed
7 years ago
1
Token Accuracy drops drastically for Tedlium database training
#112
razor1179
closed
7 years ago
0
Using dropout makes training very slow
#111
bmilde
opened
7 years ago
1
hardware setup and time taken?
#110
codegeniur
closed
7 years ago
1
training use sort_by_len false cause training speed suddenly drop
#109
Seventhen
closed
7 years ago
2
When can we enjoy the support for thchs30 corpus?
#108
colin1988
closed
7 years ago
19
How to use my Chinese corpus to train a ASR model with the example that is ars_egs/ hkust/v1
#107
Sundy1219
opened
7 years ago
7
Improved optimization and implementation of adaptive learning rate algorithms (Adagrad, RMSProp)
#106
bmilde
closed
7 years ago
19
wrap Eesen as library and expose to Kaldi
#105
xfwu
closed
7 years ago
3
Does eesen convient to extract MFCC and fbank with first order difference?
#104
zhangjiulong
closed
7 years ago
1
Problems using lexiconp.txt (lexicon with pronounciation probabilities)
#103
bmilde
opened
7 years ago
1
Improved debug messages for 'FST was not functional' problems.
#102
bmilde
closed
7 years ago
1
Added a frames-per-sec option to train-ctc-parallel
#101
bmilde
closed
7 years ago
3
How can I get the number of model parameter?
#100
dinoSpeech
opened
7 years ago
1
How to get the the sequence of phones after the acoustic model runned
#99
zhangjiulong
closed
7 years ago
4
does eesen added blanks between target labels?
#98
zhangjiulong
closed
7 years ago
1
what is the difference between tokens.txt and units.txt
#97
zhangjiulong
closed
7 years ago
4
where is the code to generate the prior of states
#96
xfwu
closed
7 years ago
2
does eesen support triphone?
#95
xfwu
closed
7 years ago
2
Dimension mistash when decoding
#94
zhangjiulong
closed
7 years ago
5
alignment generation
#93
yajiemiao
closed
7 years ago
0
force alignment
#92
xfwu
closed
7 years ago
3
sigmoid
#91
double22a
closed
7 years ago
1
const vs. constexpr issue in gpucompute/ctc-utils.h
#90
riebling
closed
7 years ago
1
Decode with avg model
#89
behnamasefi
closed
7 years ago
0
Gentle + Eesen for forced alignment
#88
migueljette
closed
7 years ago
11
Token Accuracy Drops Obj(log[Pzx])=nan
#87
behnamasefi
opened
7 years ago
11
how to prevent ""Epsilon loops exist in your decoding"
#86
xfwu
closed
7 years ago
0
Add support for CUDA toolkits 7_*, 8_* together with new CUDA compute capabilities
#85
standy66
closed
7 years ago
2
live mode recognition
#84
Bhee
closed
7 years ago
1
out of vocabulary detection
#83
madhavsund
closed
7 years ago
6
does eesen has plan for SMBR?
#82
xfwu
closed
7 years ago
3
I runned eesen on gridengine cluster only feature extraction and decoding runned on the cluster
#81
zhangjiulong
opened
7 years ago
7
setting output neurons
#80
madhavsund
closed
7 years ago
3
cluster runns ok one one node and error on two or more nodes.
#79
zhangjiulong
closed
7 years ago
4
eesen runns on gridengine qstat shows Eqs
#78
zhangjiulong
closed
7 years ago
15
how does fst prevent from generating shorter path
#77
xfwu
closed
7 years ago
2
optical character recognition -reg
#76
madhavsund
closed
8 years ago
2
setting the parameters
#75
madhavsund
closed
7 years ago
4
scoring with confidence and word insertion panelty
#74
yajiemiao
closed
8 years ago
0
Token Accuracy drops drastically for Tedlium database training
#73
razor1179
closed
7 years ago
6
Previous
Next