Neural Network Bottleneck Features for Language Identification

By Pavel Matejka, Tim Ng, Ondrej Glembek, Jeff Z. Ma, Bing Zhang, Sri Harish Reddy Mallidi. 2014

142 citations recorded by the source corpus.

This paper presents the application of Neural Network Bottleneck (BN) features in Language Identification (LID). BN f eatures are generally used for Large Vocabulary Speech Recognition in conjunction with conventional acoustic features, s uch as MFCC or PLP. We compare the BN features to several common types of acoustic features used in the state-of-the-art LID systems. 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 hig hly degraded 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 resp ect to our single best acoustic features. Index Terms: language identification, noisy speech, robust feature extraction

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Spoken Language Recognition using X-vectors (2018, methodology): “This is generally done in one of two ways: either posteriors from the ASR DNN replace those from a Gaussian mixture model (GMM) or the i-vector system is trained on bottleneck features extracted from the DNN [2, 3, 4, 5, 6].”
A Semisupervised Approach for Language Identification based on Ladder Networks (2016, methodology): “DNN have been trained for a different purpose to learn frame-level features that were then used to train a secondary classifier for the intended language recognition task [2][3].”
Approaches to Multi-domain Language Recognition (2018, methodology): “These were concatenated with PNCC features that had been contextualized with principal component analysisdiscrete cosine transform (pca-DCT) [12] to form the 102-dimensional input to the senone DNNs.”
Advances in deep neural network approaches to speaker recognition (2015, methodology): “For both the DNN/i-vector framework and the extraction of BN features, a DNN must first be trained.”
A unified deep neural network for speaker and language recognition (2015, methodology): “The impressive gains in performance obtained using deep neural networks (DNNs) for automatic speech recognition (ASR) [1] have motivated the application of DNNs to other speech technologies such as speaker recognition (SR) and language recognition (LR) [2, 3, 4, 5, 6, 7, 8, 9, 10].”
Speaker Invariant Feature Extraction for Zero-Resource Languages with Adversarial Learning (2018, methodology): “This method is widely used in various applications such as phoneme recognition [15], speaker recognition [15], and language recognition [16, 17].”