Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

speech-3

Text-Independent Speaker Verification

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in text-independent-speaker-verification-3

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

Use these libraries to find text-independent-speaker-verification-3 models and implementations

Jungjee/RawNet

2 papers 329

Datasets

No datasets available.

Subtasks

No subtasks available.

Most implemented papers

Utterance-level Aggregation for Speaker Recognition in the Wild

Andrew Zisserman, Weidi Xie, Joon Son Chung, Arsha Nagrani•Mon Feb 25 2019

This paper proposes a powerful speaker recognition deep network, using a ‘thin-ResNet’ trunk architecture, and a dictionary-based NetVLAD or GhostVLAD layer to aggregate features across time, that can be trained end-to-end.

364

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

Paper Graph

Text-Independent Speaker Verification Using 3D Convolutional Neural Networks

A. Torfi, N. Nasrabadi, J. Dawson•Thu May 25 2017

An adaptive feature learning by utilizing the 3D-CNN s for direct speaker model creation in which, for both development and enrollment phases, an identical number of spoken utterances per speaker is fed to the network for representing the speakers' utterances and creation of the speaker model.

67 0

Paper Graph

RawNet: Advanced end-to-end deep neural network using raw waveforms for text-independent speaker verification

Jee-weon Jung, Hee-Soo Heo, Ju-ho Kim, Hye-jin Shim, Ha-jin Yu•Tue Apr 16 2019

This study proposes an end-to-end system that comprises two deep neural networks, one front-end for utterance-level speaker embedding extraction and the other for back-end classification that achieves state-of-the-art performance among systems without data augmentation.

154 0

Paper Graph

Noise Invariant Frame Selection: A Simple Method to Address the Background Noise Problem for Text-independent Speaker Verification

M. Valstar, Björn Schuller, Siyang Song, L. Shen, Shuimei Zhang•Wed May 02 2018

Experiments showed that the NIFS can significantly improve the performance of Vector Quantization (VQ), Gaussian Mixture Model-Universal Background Model (GMM-UBM) and i-vector-based speaker verification systems in different unknown noisy environments with different SNRs, in comparison to their baselines.

7 0

Paper Graph

Deep multi-metric learning for text-independent speaker verification

Xinggang Wang, Wenyu Liu, Bin Feng, Jiwei Xu•Thu Jul 16 2020

Deep multi-metric learning is used to address the purpose of text-independent speaker verification and introduces three different losses for this problem, i.e., triplet loss, n-pair loss and angular loss, which work in a cooperative way to train a feature extraction network equipped with Residual connections and squeeze-and-excitation attention.

26 0

Paper Graph

Masked Proxy Loss for Text-Independent Speaker Verification

B. Raj, Jiachen Lian, Aiswarya Vinod Kumar, Hira Dhamyal, Rita Singh•Sun Nov 08 2020

A Masked Proxy (MP) loss which directly incorporates both proxy- based relationships and pair-based relationships is proposed to leverage the hardness of speaker pairs and state-of-the-art Equal Error Rate (EER) is proposed.

2 0

Paper Graph

Self-Supervised Text-Independent Speaker Verification Using Prototypical Momentum Contrastive Learning

Chao Weng, Dong Yu, Wei Xia, Chunlei Zhang, Meng Yu•Sat Dec 12 2020

A simple contrastive learning approach (SimCLR) with a momentum contrastive (MoCo) learning framework, where the MoCo speaker embedding system utilizes a queue to maintain a large set of negative examples, is examined.

91 0

Paper Graph

Y-Vector: Multiscale Waveform Encoder for Speaker Embedding

Z. Duan, Ge Zhu, Fei Jiang•Fri Oct 23 2020

A novel multi-scale waveform encoder that uses three convolution branches with different time scales to compute speech features from the waveform to outperform existing raw-waveform-based speaker embeddings on speaker verification by a large margin is proposed.

26 0

Paper Graph

Adding a benchmark result helps the community track progress.

Text-Independent Speaker Verification | State-of-the-Art