13 datasets found

Speaker verification using bidirectional multiscale feature aggregation and attentional fusion module

The ASVspoof2019 LA subset consists of three parts, training, development, and evaluation. Each partition has a disjoint set of speakers. The average duration of the utterances...

The ASVspoof 2021 dataset is a large-scale public dataset for speaker verification and spoofing countermeasures. The dataset contains various types of audio files, including...

PLDA for speaker verification with utterances of arbitrary duration.

ASVspoof 2015 and ASVspoof 2017 datasets are used for automatic speaker verification and spoofing detection. The datasets contain genuine and spoofed speech samples for logical...

The ASVspoof 2019 dataset is a large-scale public dataset for speaker verification and spoofing countermeasures. The dataset contains various types of audio files, including...

The NIST-SRE 2016 dataset contains recordings of speakers from different domains.

The Ted Talks dataset is a speaker recognition dataset, consisting of audio recordings of speakers.

Training speaker-discriminative and robust speaker verification systems without explicit speaker labels remains a persisting challenge. In this paper, we propose a new...

The ASVspoof 2021 LA dataset is used for testing the generalization capabilities of our model.

The ASVspoof 2019 LA dataset encompasses three types of speaker representation: d-vector, one-hot embedding, and VAE.

Speaker verification systems experience significant performance degradation when tasked with short-duration trial recordings. To address this challenge, a multi-scale feature...

Speaker recognition aims to identify speaker information from input speech. A type of speaker recognition is speaker verification (SV). It determines whether the test speaker's...