کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
6961179 | 1452034 | 2015 | 11 صفحه PDF | دانلود رایگان |
عنوان انگلیسی مقاله ISI
Reducing computation in an i-vector speaker recognition system using a tree-structured universal background model
دانلود مقاله + سفارش ترجمه
دانلود مقاله ISI انگلیسی
رایگان برای ایرانیان
کلمات کلیدی
موضوعات مرتبط
مهندسی و علوم پایه
مهندسی کامپیوتر
پردازش سیگنال
پیش نمایش صفحه اول مقاله
![عکس صفحه اول مقاله: Reducing computation in an i-vector speaker recognition system using a tree-structured universal background model Reducing computation in an i-vector speaker recognition system using a tree-structured universal background model](/preview/png/6961179.png)
چکیده انگلیسی
The majority of state-of-the-art speaker recognition systems (SR) utilize speaker models that are derived from an adapted universal background model (UBM) in the form of a Gaussian mixture model (GMM). This is true for GMM supervector systems, joint factor analysis systems, and most recently i-vector systems. In all of these systems, the posterior probabilities and sufficient statistics calculations represent a computational bottleneck in both enrollment and testing. We propose a multi-layered hash system, employing a tree-structured GMM-UBM which uses Runnalls' Gaussian mixture reduction technique, in order to reduce the number of these calculations. With this tree-structured hash, we can trade-off reduction in computation with a corresponding degradation of equal error rate (EER). As an example, we reduce this computation by a factor of 15Ã while incurring less than 10% relative degradation of EER (or 0.3% absolute EER) when evaluated with NIST 2010 speaker recognition evaluation (SRE) telephone data.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Speech Communication - Volume 66, February 2015, Pages 36-46
Journal: Speech Communication - Volume 66, February 2015, Pages 36-46
نویسندگان
Richard McClanahan, Phillip L. De Leon,