Approaches for Language Identification in Mismatched Environments
Abstract
In this paper, we consider the task of language identification in the context of mismatch conditions. Specifically, we address the issue of using unlabeled data in the domain of interest to improve the performance of a state-of-the-art system. The evaluation is performed on a 9-language set that includes data in both conversational telephone speech and narrowband broadcast speech. Multiple experiments are conducted to assess the performance of the system in this condition and a number of alternatives to ameliorate the drop in performance. The best system evaluated is based on deep neural network bottleneck features using i-vectors. The proposed system results in a 30% improvement over the baseline result.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 08, 2016
- Accession Number
- AD1033826
Entities
People
- Gabriel Martinez-montes
- Pedro A. Torres-carrasquillo
- Shahan C. Nercessian
Organizations
- MIT Lincoln Laboratory