Title | ||
---|---|---|
S-vector: A discriminative representation derived from i-vector for speaker verification |
Abstract | ||
---|---|---|
Representing data in ways to disentangle and factor out hidden dependencies is a critical step in speaker recognition systems. In this work, we employ deep neural networks (DNN) as a feature extractor to disentangle and emphasize the speaker factors from other sources of variability in the commonly used i-vector features. Denoising autoencoder based unsupervised pre-training, random dropout fine-tuning, and Nesterov accelerated gradient based momentum is used in DNN training. Replacing the i-vectors with the resulting speaker vectors (s-vectors), we obtain superior results on NIST SRE corpora on a wide range of operating points using probabilistic linear discriminant analysis (PLDA) back-end. |
Year | Venue | Keywords |
---|---|---|
2015 | European Signal Processing Conference | speaker verification,denoising autoencoder,random dropout |
Field | DocType | ISSN |
Noise reduction,Pattern recognition,Noise measurement,Computer science,Feature extraction,Robustness (computer science),Speech recognition,NIST,Speaker recognition,Artificial intelligence,Artificial neural network,Discriminative model | Conference | 2076-1465 |
Citations | PageRank | References |
1 | 0.35 | 13 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yusuf Ziya Isik | 1 | 1 | 1.02 |
H. Erdogan | 2 | 589 | 55.11 |
Ruhi Sarikaya | 3 | 698 | 64.49 |