2015年8月11日 (二) 08:19 Lilt

2015-08-11T08:19:21Z

Lilt：以“Weekly Summary 1. Prepare for three deep speaker embedding tasks: 1). large-scale deep speaker vector framework -- hold. 2). knowledge transfer for i-vector -- in...”为内容创建页面

2015-08-11T08:18:08Z

以“Weekly Summary 1. Prepare for three deep speaker embedding tasks: 1). large-scale deep speaker vector framework -- hold. 2). knowledge transfer for i-vector -- in...”为内容创建页面

新页面

Weekly Summary

1. Prepare for three deep speaker embedding tasks:

1). large-scale deep speaker vector framework -- hold.

2). knowledge transfer for i-vector -- in process.

The ivector knowledge transfor(IKT) both on DNN and RNN has been done.

The results show that the performance of DNN-IKT is better than former d-vector system,

but still worse than i-vector baseline. And the RNN-IKT does not work well.

We think it is due to the training-data is insufficient.

3). derive binary i-vectors using Hamming distance learning

obtain the code from Xing Chao, and start to modify the test-code and design the experiments.

Next Week

1. Go on the task 1 and repeat the DNN-IKT of 2) on Fisher and SRE database.

2. Go on the task 1 and attempt to complete 3).

@@ 第15行： / 第15行： @@
 We think it is due to the training-data is insufficient.
-). derive binary i-vectors using Hamming distance learning
+). derive binary i-vectors using Hamming distance learning (identification efficiency Exp.)
 obtain the code from Xing Chao, and start to modify the test-code and design the experiments.

Lantian Li 15-08-11 - 版本历史

2015年8月11日 (二) 08:19 Lilt

Lilt：以“Weekly Summary 1. Prepare for three deep speaker embedding tasks: 1). large-scale deep speaker vector framework -- hold. 2). knowledge transfer for i-vector -- in...”为内容创建页面