REVIEW 2 cited by
In defence of metric learning for speaker recognition
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
The objective of this paper is 'open-set' speaker recognition of unseen speakers, where ideal embeddings should be able to condense information into a compact utterance-level representation that has small intra-speaker and large inter-speaker distance. A popular belief in speaker recognition is that networks trained with classification objectives outperform metric learning methods. In this paper, we present an extensive evaluation of most popular loss functions for speaker recognition on the VoxCeleb dataset. We demonstrate that the vanilla triplet loss shows competitive performance compared to classification-based losses, and those trained with our proposed metric learning objective outperform state-of-the-art methods.
Forward citations
Cited by 2 Pith papers
-
Non-Adaptive Adversarial Face Generation
A non-adaptive batch of 100 face queries against a face recognition API can produce synthetic faces that impersonate a target identity with a chosen attribute, exceeding 93% success against AWS CompareFaces at its def...
-
Addressing malware family concept drift with triplet autoencoder
Combining a triplet autoencoder with DBSCAN centroid thresholds improves detection of unseen malware families in two datasets, but the temporal evaluation protocol and unreported hyperparameters undermine the results.
Discussion (0). Sign in to comment.