Generative X-Vectors for Text-Independent Speaker Verification

Xu, Longting; Das, Rohan Kumar; Yılmaz, Emre; Yang, Jichen; Li, Haizhou

doi:10.1109/slt.2018.8639510

Cited by 11 publications

(8 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experimental settings follow our previous work [4]. The baseline systems are i‐vector and x‐vector based systems, respectively.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Linear transformation on x‐vector for text‐independent speaker verification

Ren²,

Zhang

et al. 2019

Electron. lett.

Self Cite

View full text Add to dashboard Cite

Total variability model-based i-vector and deep neural network-based embedding x-vector are both widely used for text-independent speaker verification. In this Letter, a novel model is proposed, which can contain information of both i-vector and x-vector by using parallel factor analysis. The authors aim to obtain a linear transformation expression for x-vectors based on background i-vectors and x-vectors, and consider the linearly transformed x-vector as the novel model, thus they name it as x l-vector. The novel x l-vector can maximise intra and minimise inner speaker variability, in addition, it can improve the system performance without latency. Experiments were conducted on NIST 2010 dataset, and in terms of equal error rate, they observe up to 37.27 and 53.38% relative improvement of the authors proposed x l-vector model compared to the i-vector and x-vector models, respectively.

show abstract

“…The experimental settings follow our previous work [4]. The baseline systems are i‐vector and x‐vector based systems, respectively.…”

Section: Resultsmentioning

confidence: 99%

“…The increase percentage will be introduced in the next section. Experimental results: The experimental settings follow our previous work [4]. The baseline systems are i-vector and x-vector based systems, respectively.…”

mentioning

confidence: 99%

Linear transformation on x‐vector for text‐independent speaker verification

Ren²,

Zhang

et al. 2019

Electron. lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…To get the X-Vectors, we extract embedded vectors independently on the text using a pre-trained model 1 . As stated before, this model was initially trained for a speaker verification task [8,9,10] using NIST SRE recipe supported in the Kaldi toolkit. The details about the recipe and the pretained model are available in author's github 2 .…”

Section: X-vectormentioning

confidence: 99%

Introducing Prosodic Speaker Identity for a Better Expressive Speech Synthesis Control

Sini¹,

Maguer²,

Lolive³

et al. 2020

Speech Prosody 2020

View full text Add to dashboard Cite

To have more control over Text-to-Speech (TTS) synthesis and to improve expressivity, it is necessary to disentangle prosodic information carried by the speaker's voice identity from the one belonging to linguistic properties. In this paper, we propose to analyze how information related to speaker voice identity affects a Deep Neural Network (DNN) based multi-speaker speech synthesis model. To do so, we feed the network with a vector encoding speaker information in addition to a set of basic linguistic features. We then compare three main speaker coding configurations: a) simple one-hot vector describing the speaker gender and identifier ; b) an embedding vector extracted from a speaker recognition pre-trained model ; c) a prosodic vector which summarizes information such as melody, intensity, and duration. To measure the impact of the input feature vector, we investigate the representation of the latent space at the output of the first layer of the network. The aim is to have an overview of our data representation and model behavior. Furthermore, we conducted a subjective assessment to validate the result. Results show that the prosodic identity of the speaker is captured by the model and therefore allows the user to control more precisely synthesis.

show abstract

“…By introducing deep learning into anti-spoofing, deep neural networks (DNN) have achieved promising results in anti-spoofing of ASVspoof 2017 [9]- [11] and ASVspoof 2019 [12]. ASV as a standalone task has also gained great improvement from deep learning [13], [14]. Given those achievements and in order to make ASV and anti-spoofing a step forward to practical usage, some early studies have proposed that a separately designed antispoofing system is implement before ASV, only the utterances which have passed spoofing detection are verified again by a ASV system [4], [14], which is in fact a cascaded structure as is illustrated in FIGURE 1.…”

Section: Introductionmentioning

confidence: 99%

Joint Decision of Anti-Spoofing and Automatic Speaker Verification by Multi-Task Learning With Contrastive Loss

Sun

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Automatic speaker verification (ASV) is an emerging biometric verification technique with more and more applications. However, both verification accuracy and anti-spoofing should be considered carefully before putting ASV into practice, where anti-spoofing is also called replay detection in which voice is recorded, stored and replayed to deceive ASV systems. Cascaded decision of anti-spoofing and ASV is a straightforward solution to tackle the two issues. In this paper, joint decision of anti-spoofing and ASV was investigated in a multi-task learning framework with contrastive loss in order to improve the cascaded decision approach. A modified triplet loss was firstly constructed to supervise deep neural networks to extract embedding vectors containing information of both speaker identity and spoofing. The embedding vectors were subsequently taken as input features by back-end classifiers towards speaker and spoofing classification. The experimental results on both ASVspoof 2017 and ASVspoof 2019 showed that the proposed joint decision approach with triplet loss outperformed the corresponding baselines, a recent work on joint decision with Gaussian back-end fusion and our previous joint decision approach with cross-entropy loss.

show abstract

Generative X-Vectors for Text-Independent Speaker Verification

Cited by 11 publications

References 29 publications

Linear transformation on x‐vector for text‐independent speaker verification

Linear transformation on x‐vector for text‐independent speaker verification

Introducing Prosodic Speaker Identity for a Better Expressive Speech Synthesis Control

Joint Decision of Anti-Spoofing and Automatic Speaker Verification by Multi-Task Learning With Contrastive Loss

Contact Info

Product

Resources

About