Neural Network Bottleneck Features for Language Identification

This paper presents the application of Neural Network Bot-tleneck (BN) features in Language Identification (LID). BN fea-tures are generally used for Large Vocabulary Speech Recogni-tion in conjunction with conventional acoustic features, such as MFCC or PLP. We compare the BN features to several common types of acoustic features used in the state-of-the-art LID sys-tems. The test set is from DARPA RATS (Robust Automatic Transcription of Speech) program, which seeks to advance state-of-the-art detection capabilities on audio from highly de-graded radio communication channels. On this type of noisy data, we show that in average, the BN features provide a 45% relative improvement in the Cavgor Equal Error Rate (EER) metrics across several test duration conditions, with respect to our single best acoustic features. Index Terms: language identification, noisy speech, robust feature extraction