“2013-04-19”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
第9行: 第9行:
 
  (1) 400 hour BN model.  
 
  (1) 400 hour BN model.  
  
  (2) Tencent test result: 70h training data(2 day, 15 machines, 10 threads), 88k LM, general test case:  
+
  (2) Tencent test result: 70h training data(2 day, 15 machines, 10 threads),  
 +
88k LM, general test case:  
  
 
gmmi-bmmi: 38.7%
 
gmmi-bmmi: 38.7%

2013年4月19日 (五) 06:25的版本

1. Data sharing

(1) AM/lexicon/LM are shared.
(2) LM count files are still in transfering. 

2. DNN progress

(1) 400 hour BN model. 
(2) Tencent test result: 70h training data(2 day, 15 machines, 10 threads), 

88k LM, general test case:

gmmi-bmmi: 38.7% dnn-1: 28% 11 frame window, phone-based tree dnn-2: 34% 9 frame window, state-based tree

(3) GPU & CPU merge. Invesigate the possibility to merge GPU and 

CPU code. Try to find out an easier way. (1 week)

(4) L-1 sparse initial training. 

3.Kaldi/HTK merge

(1) HTK2Kaldi: the tool with Kaldi does not work.
(2) Kaldi2HTK: done with implementation. Testing?

4. Embedded progress

(1). Some large performance (speed) degradation with the embedded platform(1/60).

(2). Planning for sparse DNN.

(3). QA LM training, Mengyuan?