In this paper, we use data augmentation to improve performance of deep neural network (DNN) embeddings for speaker recognition. The DNN, which is trained to discriminate between speakers, maps variable-length utterances to fixed-dimensional embeddings that we call x-vectors. Prior studies have found that embeddings leverage large-scale training datasets better than i-vectors. However, it can be challenging to collect substantial quantities of labeled data for training. We use data augmentation, consisting of added noise and reverberation, as an inexpensive method to multiply the amount of training data and improve robustness. The x-vectors are compared with i-vector baselines on Speakers in the Wild and NIST SRE 2016 Cantonese. We find that while augmentation is beneficial in the PLDA classifier, it is not helpful in the i-vector extractor. However, the x-vector DNN effectively exploits data augmentation, due to its supervised training. As a result, the x-vectors achieve superior performance on the evaluation datasets.

X-Vectors: Robust DNN Embeddings for Speaker Recognition

From the Publisher:
Andrew Tanenbaum and Maarten van Steen cover the principles, advanced concepts, and technologies of distributed systems in detail, including: communication, replication, fault tolerance, and security. Intended for use in a senior/graduate level distributed systems course or by professionals, this text systematically shows how distributed systems are designed and implemented in real systems. Written in the superb writing style of other Tanenbaum books, the material also features unique accessibility and a wide variety of real-world examples and case studies, such as NFS v4, CORBA, DOM, Jini, and the World Wide Web.

FEATURES

Detailed coverage of seven key principles.
An introductory chapter followed by a chapter devoted to each key principle: communication, processes, naming, synchronization, consistency and replication, fault tolerance, and security, including unique comprehensive coverage of middleware models.
Four chapters devoted to state-of-the-art real-world examples of middleware. Covers object-based systems, document-based systems, distributed file systems, and coordination-based systems including CORBA, DCOM, Globe, NFS v4, Coda, the World Wide Web, and Jini.
Excellent coverage of timely, advanced, distributed systems topics: Security, payment systems, recent Internet and Web protocols, scalability, and caching and replication.
NEW-The Prentice Hall Companion Website for this book contains PowerPoint slides, figures in various file formats, and other teaching aids, and a link to the author's Web site.

/pdf/distributed-systems-principles-and-paradigms-1f0v69ldk2.pdf

Distributed Systems: Principles and Paradigms

Latent semantic models, such as LSA, intend to map a query to its relevant documents at the semantic level where keyword-based matching often fails In this study we strive to develop a series of new latent semantic models with a deep structure that project queries and documents into a common low-dimensional space where the relevance of a document given a query is readily computed as the distance between them The proposed deep structured semantic models are discriminatively trained by maximizing the conditional likelihood of the clicked documents given a query using the clickthrough data To make our models applicable to large-scale Web search applications, we also use a technique called word hashing, which is shown to effectively scale up our semantic models to handle large vocabularies which are common in such tasks The new models are evaluated on a Web document ranking task using a real-world data set Results show that our best model significantly outperforms other latent semantic models, which were considered state-of-the-art in the performance prior to the work presented in this paper

/pdf/learning-deep-structured-semantic-models-for-web-search-3ouh9c4skm.pdf

Learning deep structured semantic models for web search using clickthrough data

http://cs.uef.fi/sipu/pub/SRE-review-Kinnunen-and-Li.pdf

An overview of text-independent speaker recognition: From features to supervectors

In this paper we investigate the use of deep neural networks (DNNs) for a small footprint text-dependent speaker verification task. At development stage, a DNN is trained to classify speakers at the frame-level. During speaker enrollment, the trained DNN is used to extract speaker specific features from the last hidden layer. The average of these speaker features, or d-vector, is taken as the speaker model. At evaluation stage, a d-vector is extracted for each utterance and compared to the enrolled speaker model to make a verification decision. Experimental results show the DNN based speaker verification system achieves good performance compared to a popular i-vector system on a small footprint text-dependent speaker verification task. In addition, the DNN based system is more robust to additive noise and outperforms the i-vector system at low False Rejection operating points. Finally the combined system outperforms the i-vector system by 14% and 25% relative in equal error rate (EER) for clean and noisy conditions respectively.

/pdf/deep-neural-networks-for-small-footprint-text-dependent-1otxqwx3q4.pdf

Deep neural networks for small footprint text-dependent speaker verification

https://www.microsoft.com/en-us/research/wp-content/uploads/2000/01/1-s2.0-S0167639399000771-main.pdf

Robustness to telephone handset distortion in speaker recognition by discriminative feature design

This paper describes a new approach to acoustic modeling for large vocabulary continuous speech recognition (LVCSR) systems. Each phone is modeled with a large Gaussian mixture model (GMM) whose context-dependent mixture weights are estimated with a sentence-level discriminative training criterion. The estimation problem is cast in a neural network framework, which enables the incorporation of the appropriate constraints on the mixture weight vectors, and allows a straight-forward training procedure, based on steepest descent. Experiments conducted on the Callhome-English and Switchboard databases show a significant improvement of the acoustic model performance, and a somewhat lesser improvement with the combined acoustic and language models.

Mitch Weintraub

Papers

Robustness to telephone handset distortion in speaker recognition by discriminative feature design

Discriminative mixture weight estimation for large Gaussian mixture models