English
Related papers

Related papers: Max-margin Metric Learning for Speaker Recognition

200 papers

We present an approach to tackle the speaker recognition problem using Triplet Neural Networks. Currently, the $i$-vector representation with probabilistic linear discriminant analysis (PLDA) is the most commonly used technique to solve…

Sound · Computer Science 2019-10-07 Kin Wai Cheuk , Balamurali B. T. , Gemma Roig , Dorien Herremans

Probabilistic linear discriminant analysis (PLDA) has broad application in open-set verification tasks, such as speaker verification. A key concern for PLDA is that the model is too simple (linear Gaussian) to deal with complicated data;…

Sound · Computer Science 2021-11-25 Di Wang , Lantian Li , Hongzhi Yu , Dong Wang

In this work, a novel solution to the speaker identification problem is proposed through minimization of statistical divergences between the probability distribution (g). of feature vectors from the test utterance and the probability…

Machine Learning · Statistics 2015-12-17 Ayanendranath Basu , Smarajit Bose , Amita Pal , Anish Mukherjee , Debasmita Das

State-of-the-art i-vector based speaker verification relies on variants of Probabilistic Linear Discriminant Analysis (PLDA) for discriminant analysis. We are mainly motivated by the recent work of the joint Bayesian (JB) method, which is…

Sound · Computer Science 2017-01-20 Yiyan Wang , Haotian Xu , Zhijian Ou

Acoustic models using probabilistic linear discriminant analysis (PLDA) capture the correlations within feature vectors using subspaces which do not vastly expand the model. This allows high dimensional and correlated feature spaces to be…

Computation and Language · Computer Science 2015-06-23 Liang Lu , Steve Renals

PLDA is a popular normalization approach for the i-vector model, and it has delivered state-of-the-art performance in speaker verification. However, PLDA training requires a large amount of labeled development data, which is highly…

Sound · Computer Science 2016-09-28 Chenghui Zhao , Lantian Li , Dong Wang , April Pu

Traditional speech separation and speaker diarization approaches rely on prior knowledge of target speakers or a predetermined number of participants in audio signals. To address these limitations, recent advances focus on developing…

Recently, speaker embeddings extracted from a speaker discriminative deep neural network (DNN) yield better performance than the conventional methods such as i-vector. In most cases, the DNN speaker classifier is trained using cross entropy…

Audio and Speech Processing · Electrical Eng. & Systems 2019-06-19 Xu Xiang , Shuai Wang , Houjun Huang , Yanmin Qian , Kai Yu

This article presents a novel approach for learning domain-invariant speaker embeddings using Generative Adversarial Networks. The main idea is to confuse a domain discriminator so that is can't tell if embeddings are from the source or…

Audio and Speech Processing · Electrical Eng. & Systems 2018-11-08 Gautam Bhattacharya , Joao Monteiro , Jahangir Alam , Patrick Kenny

This paper focuses on multi-enrollment speaker recognition which naturally occurs in the task of online speaker clustering, and studies the properties of different scoring back-ends in this scenario. First, we show that popular cosine…

Audio and Speech Processing · Electrical Eng. & Systems 2023-02-21 Alexey Sholokhov , Nikita Kuzmin , Kong Aik Lee , Eng Siong Chng

Probabilistic Linear Discriminant Analysis (PLDA) was the dominant and necessary back-end for early speaker recognition approaches, like i-vector and x-vector. However, with the development of neural networks and margin-based loss…

Sound · Computer Science 2022-04-26 Zhuo Li , Runqiu Xiao , Zihan Zhang , Zhenduo Zhao , Wenchao Wang , Pengyuan Zhang

This paper presents a linear regression based back-end for speaker verification. Linear regression is a simple linear model that minimizes the mean squared estimation error between the target and its estimate with a closed form solution,…

Sound · Computer Science 2018-02-13 Xiao-Lei Zhang

The x-vector maps segments of arbitrary duration to vectors of fixed dimension using deep neural network. Combined with the probabilistic linear discriminant analysis (PLDA) backend, the x-vector/PLDA has become the dominant framework in…

Audio and Speech Processing · Electrical Eng. & Systems 2020-01-15 Bin Gu , Wu Guo

We investigate deep neural network performance in the textindependent speaker recognition task. We demonstrate that using angular softmax activation at the last classification layer of a classification neural network instead of a simple…

Sound · Computer Science 2018-04-27 Sergey Novoselov , Andrey Shulipa , Ivan Kremnev , Alexandr Kozlov , Vadim Shchemelinin

In diarization, the PLDA is typically used to model an inference structure which assumes the variation in speech segments be induced by various speakers. The speaker variation is then learned from the training data. However, human…

Audio and Speech Processing · Electrical Eng. & Systems 2020-09-01 Jiamin Xie , Suzanna Sia , Paola Garcia , Daniel Povey , Sanjeev Khudanpur

We present MooseNet, a trainable speech metric that predicts the listeners' Mean Opinion Score (MOS). We propose a novel approach where the Probabilistic Linear Discriminative Analysis (PLDA) generative model is used on top of an embedding…

Computation and Language · Computer Science 2023-10-27 Ondřej Plátek , Ondřej Dušek

The conventional speaker recognition frameworks (e.g., the i-vector and CNN-based approach) have been successfully applied to various tasks when the channel of the enrolment dataset is similar to that of the test dataset. However, in…

Audio and Speech Processing · Electrical Eng. & Systems 2019-02-26 Xin Fang , Liang Zou , Jin Li , Lei Sun , Zhen-Hua Ling

Probabilistic linear discriminant analysis (PLDA) has been widely used in open-set verification tasks, such as speaker verification. A potential issue of this model is that the training set often contains limited number of classes, which…

Sound · Computer Science 2021-11-25 Jiao Han , Yunqi Cai , Lantian Li , Guanyu Li , Dong Wang

In this paper, we propose an effective training strategy to ex-tract robust speaker representations from a speech signal. Oneof the key challenges in speaker recognition tasks is to learnlatent representations or embeddings containing…

Audio and Speech Processing · Electrical Eng. & Systems 2020-08-05 Yoohwan Kwon , Soo-Whan Chung , Hong-Goo Kang

Speaker embedding extractors are typically trained using a classification loss over the training speakers. During the last few years, the standard softmax/cross-entropy loss has been replaced by the margin-based losses, yielding significant…

Audio and Speech Processing · Electrical Eng. & Systems 2024-06-19 Themos Stafylakis , Anna Silnova , Johan Rohdin , Oldrich Plchot , Lukas Burget