Accession Number:

AD1033667

Title:

The MITLL NIST LRE 2015 Language Recognition System

Descriptive Note:

Technical Report

Corporate Author:

MIT Lincoln Laboratory Lexington United States

Personal Author(s):

• Sturim,Douglas E
• Richardson,Frederick S
• Singer,Elliot
• Godoy,Elizabeth C
• Reynolds,Douglas A
• Torres-Carrasquillo,Pedro A
• Dehak,Najim
• Shum,Stephen

Report Date:

2016-05-06

Pagination or Media Count:

8.0

Abstract:

In this paper we describe the most recent MIT Lincoln Laboratory language recognition system developed for the NIST 2015 Language Recognition Evaluation LRE. The submission features a fusion of five core classifiers, with most systems developed in the context of an i-vector framework. The 2015 evaluation presented new paradigms. First, the evaluation included fixed training and open training tracks for the first time second, language classification performance was measured across 6 language clusters using 20 language classes instead of an N-way language task and third, performance was measured across a nominal 3-30 second range. Results are presented for the overall performance across the six language clusters for both the fixed and open training tasks. On the 6-cluster metric the Lincoln system achieved overall costs of 0.173 and 0.168 for the fixed and open tasks respectively.

Descriptors:

• AUTOMATED SPEECH RECOGNITION
• monte carlo method
• discriminant analysis
• digital data
• pattern recognition
• classification
• artificial neural networks
• statistical analysis
• normal distribution
• order statistics
• algorithms
• natural language
• DEEP LEARNING

Subject Categories:

• Voice Communications
• Statistics and Probability

Distribution Statement:

APPROVED FOR PUBLIC RELEASE