利用語者特定背景模型之語者確認系統

在一般的語者確認系統中，有兩種背景語者模型的選取方法，分別為通用背景模型(Universal Background Model,UBM)與反語者模型(Anti-SpeakerModel)，其兩種方法都存在著各自的缺點，將會使系統的效能下降。所以在本論文中我們將對背景語者模型進行研究與改良，主要以改善以上兩種背景語者模型之缺點、提升辨識效能為目的，並提出背景語者模型之新準則，以及找出一個符合此準則的目標函式，然後將每個由通用背景模型所調適出來的語者特定模型，分別建立其專屬的背景語者模型，在本論文中我們稱這些符合新準則的模型為語者特定背景模型(Speaker Dependent Background Model, SDBM)。語者特定背景模型將可改善傳統反語者模型與通用背景模型的部分缺點，並增進語者確認的效果，其效果將以實驗來予以驗證。

關鍵字

語者確認；通用背景模型；反語者模型

並列摘要

Universal background model (UBM) and anti-speaker model are two methods of background models for a speaker verification system in general. But they existed a few problems. Therefore we propose two criteria for determining background models. The created new background model is called the speaker dependent background model (SDBM). The results of experiments show that the SDBM improves the performance of the UBM and anti-speaker model approaches.

並列關鍵字

Speaker Verification ； Universal Background Model ； Anti-Model

參考文獻

[3]J. L. Gauvain and C. H. Lee, “Maximum a Posteriori Estimation for Multivariate Gaussian Mixture Observations of Markov Chains,” IEEE Transactions on Speech and Audio Processing, vol.2, pp. 291-298, 1994.

[4]D.A. Reynolds, T.F. Quatieri, and R.B. Dunn, “Speaker Verification Using Adapted Gaussian Mixture Models,” Digital Signal Processing, vol.10, pp. 19-41, 2000.

[5]B.H. Juang and S. Katagiri, “Discriminative Learning for Minimum Error Classification [Pattern Recognition],” Signal Processing, IEEE Transactions on, vol.40, pp.3043-3054, 1992.

[7]P. Kenny, G. Boulianne, P. Ouellet and P. Dumouchel, “Joint Factor Analysis Versus Eigenchannels in Speaker Recognition,” Audio, Speech, and Language Processing, IEEE Transactions on, vol.15, pp.1435-1447, 2007.

[9]B. Yegnanarayana and S. P. Kishore, “AANN: An Alternative to GMM for Pattern Recognition,” Neural Network, vol.15, pp. 459–469, 2002.

國際替代計量

利用語者特定背景模型之語者確認系統

未授權

主題瀏覽