A fast Speaker Identification method using nearest neighbor distance
Authors:
- 1 October 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3 (21645221) , 2159-2162
Abstract
By increasing the number of registered speakers in Speaker Identification (SI) systems, computation time for identifying an unknown speaker is significantly increased. This problem arises from the simple design of conventional methods. Due to this limitation, we cannot use conventional SI methods in real time applications. In this paper, we propose a two-step method to overcome this limitation. We use different identification methods for each step. In the first step we reduce the search space using Nearest Neighbor method. In the second step we identify the target speaker using the conventional GMM-based SI method. The experimental results show 3.4× speed-ups without any accuracy loss using the proposed method. It is also shown that if the number of selected speakers in first step is furthermore reduced, the identification speed will increase, at the cost of a slight decrease of accuracy. So, there is a trade-off between accuracy and speed-up that can be controlled by a threshold.Keywords
This publication has 6 references indexed in Scilit:
- A fast two-level Speaker Identification method employing sparse representation and GMM-based methods2012
- Speaker Model Clustering for Efficient Speaker Identification in Large Population ApplicationsIEEE Transactions on Audio, Speech, and Language Processing, 2009
- Hierarchical speaker identification using speaker clustering2004
- A Speaker Pruning Algorithm for Real-Time Speaker Identification2003
- Speaker Verification Using Adapted Gaussian Mixture ModelsDigital Signal Processing, 2000
- Speaker identification and verification using Gaussian mixture speaker modelsSpeech Communication, 1995