Task List - 版本历史

2016年10月16日 (日) 12:31 Tangzy

2016-10-16T12:31:07Z

2016年10月16日 (日) 12:29 Tangzy

2016-10-16T12:29:47Z

2016年10月16日 (日) 12:17 Tangzy

2016-10-16T12:17:05Z

Tangzy：以“previous”替换内容

2016-10-16T12:16:44Z

以“previous”替换内容

Zhangzy：/* Speech Recognition */

2016-02-19T07:03:24Z

‎Speech Recognition

Zhangzy：/* Network architecture test */

2016-02-19T06:47:04Z

‎Network architecture test

Zhangzy：/* CTC expanded */

2016-02-19T06:39:13Z

‎CTC expanded

Tangzy：/* Speech Recognition */

2016-02-19T06:24:44Z

‎Speech Recognition

2015年10月26日 (一) 07:59 Zhaomy

2015-10-26T07:59:08Z

Zhaomy：/* End-to-End speech recognition */

2015-10-26T07:35:31Z

‎End-to-End speech recognition

@@ 第1行： / 第1行： @@
 =Tasks at hand=
 ==Speech Recognition==
@@ 第10行： / 第9行： @@
 ===visualization===
 * Ying Shi, Zhiyuan Tang
 ==Speaker Recognition==
 *Lantian Li, Yixiang Chen
@@ 第29行： / 第25行： @@
 =Projects=
 ------------------------------
 [[task previous]]

←上一版本		2016年10月16日 (日) 12:29的版本
第1行：		第1行：
		+
		+	=Tasks at hand=
		+
		+
		+	==Speech Recognition==
		+
		+	===joint learning===
		+	* Hang Luo, Zhiyuan Tang
		+
		+	===visualization===
		+	* Ying Shi, Zhiyuan Tang
		+
		+
		+	==Speaker Recognition==
		+	*Lantian Li, Yixiang Chen
		+
		+
		+
		+
		+	=Tasks Done=
		+
		+	=Technical Reports to write=
		+
		+	=Papers to write=
		+
		+	=Patents to write=
		+
		+	=Patents done=
		+
		+	=Projects=
		+
		+
		+
		+
		+
		+
		+
		+
		+
		+
		+
		+
		+	------------------------------
		+
	[[task previous]]		[[task previous]]

←上一版本		2016年10月16日 (日) 12:17的版本
第1行：		第1行：
−	[[previous]]	+	[[task previous]]

@@ 第1行： / 第1行： @@
-=Task To Do=
+[[previous]]
-==Speech Recognition==
-−
-===CTC expanded===
-*Voice activity detection
-:*LSTM+CTC
-:*TDNN+CTC
-::* BLANK as silence, others as speech
-−
-*Keyword detection
-:* Character/Word-level, external key-word fst
-:* G-fst need to be signal word?
-−
-*Emotion recognition
-:* LSTM-CTC
-−
-===Network architecture test===
-*chain model:
-:*tdnn + simple-lstm
-::* only keep forget gate
-−
-*ctc+mpe
-:* similar to chain training
-−
-===Spiral Joint Training of SPEECH and SPEAKER===
-*ASR & SID parallel training and benefit mutual
-−
-===Small data-set and Big model===
-*Investigate the efficiency of pre-training  on small/big-model using dark-knowledge
-−
-===Low-resource language improvement===
-*SID
-:* How to improve low-resource speaker
-−
-===End-to-End speech recognition===
-* Discriminative-Learning code implementation
-:* Zhiyuan Tang
-−
-===Multi-task===
-* Fusion of speech-recognition and speech-rate
-:* Xiangyu Zeng
-* Self-informed neural network structure learning
-:* Mengyuan Zhao
-−
-===Integrate the class information to HCLG fst for speech recognition===
-*zhiyuan
-−
-===Distant speech recognition===
-*RNN-DAE: echo or reverberation
-:*Xuewei Zhang/Zhiyuan Tang/Mengyuan Zhao/Zhiyong Zhang
-*Reverberation
-:*Mutli-microphones
-:*(Lasso),Xuewei Zhang
-−
-===Voice conversation===
-*hold
-−
-===Sparse DNN===
-*Zhiyuan Tang
-−
-===Correlation based SENONE cluster===
-−
-===NN Multi-GPU parallel traing===
-*Multi-GPU using data parallelization
-:*Sheng Su
-* nnet3 mpe
-:* Xuewei Zhang
-−
-===Audio Embedding===
-*Ke Ning
-−
-===RNN training accelerating===
-−
-===Data selection===
-*Zhiyong Zhang
-*Sub-modular data selection
-*Objective-function loss training self-adaptation
-−
-===Decoder===
-*Confidence output for task-required
-−
-==Speaker Verification==
-===binary code===
-*Lantian Li
-−
-===RNN-ivector===
-*Lantian Li
-−
-===DNN clustering===
-*Lantian Li
-−
-=Task DONE=
-==Multi-Mode features based VAD==
-* Shi Yin
-−
-==DNN based Language identification and Speaker identification==
-* Xuewei Zhang/Zhiyuan Tang
-−
-==Neural network visulization==
-* Mian Wang,DONE
-−
-==Dark knowledge==
-* Mengyuan Zhao, Xiangyu Zeng, Zhiyong Zhang, Chao Liu
-−
-==Normal RNN speech recognition==
-* Mengyuan Zhao
-−
-==Monmentum-like Hessien-Free acceleration==
-* Nestrov/Adagrad/AdaDelta/AdaM
-* Zhiyong Zhang/Xiangyu Zeng
-−
-==Activation value normalization through time --Batch Normalization==
-* Zhiyong Zhang
-−
-==Mix-training Balance decision tree==
-* Zhiyong Zhang
-−
-==20-h Chinese data-set release==
-* Xuewei Zhang
-−
-==Unbound activation function(Rectifier/Maxout/Pnorm) go-through searching method==
-* nne3 test --Xuewei Zhang
-−
-=Technical Report To Write=
-, DNN-DAE based noise cancellation -- Xiangyu Zeng / Mengyuan Zhao / Zhiyong Zhang  --DONE
-, Speech Rate DNN speech recognition --Shi Yin/Xiangyu Zeng --DONE
-, CNN+fbank feature combination --Mian Wang /Yiye Lin /Mengyuan Zhao /Shi Yin
-, Uyghur low-resource acoustic model enhancement -- Shi Yin / Mengyuan Zhao / Zhiyong Zhang --DONE
-, Uyghur 20h database release --Kaer /Shi Yin --DONE
-，Dark-Knowledge Transfer
-    *: Xiangyu Zeng/ Mengyuan Zhao / Zhiyong Zhang
-−
-=Paper to Write=
-−
-=Patent done=
-* A method of new word enhancement for speech recognition --Yue Zhang
-−
-=Project=
-* Xiaomi TV
-:*Mengyuan Zhao/Zhiyong Zhang
-:*TAG-lm & Domain-specific general lm
-*Chinese-English mix-training

@@ 第22行： / 第22行： @@
 *ctc+mpe
 :* similar to chain training
 ===End-to-End speech recognition===