Probabilistic Linear Discriminant Analysis

By Sergey Ioffe. 2006

576 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

VoxCeleb: A Large-Scale Speaker Identification Dataset (2017, methodology): “The PLDA scoring function [41] is used for verification.”
The AS-NU System for the M2VoC Challenge (2021, methodology): “For Track 2, to tackle the shortage of the target utterances, based on probabilistic linear discriminant analysis (PLDA) [19], the speaker-wise x-vector of the 5 target utterances was used to select the speaker most similar to the target speaker from the AIShell3 training set.”
Self-Attentive Speaker Embeddings for Text-Independent Speaker Verification (2018, background): “The work in [16] divided the end-to-end system into two components: a DNN to produce speaker embeddings and a separately trained PLDA classifier to compare embedding pairs.”
Diarization is Hard: Some Experiences and Lessons Learned for the JHU Team in the Inaugural DIHARD Challenge (2018, methodology): “In both cases, PLDA parameters were trained with VoxCeleb, and thresholds were learned in a supervised fashion on the dev data.”
CN-Celeb: A Challenging Chinese Speaker Recognition Dataset (2019, methodology): “This data set was used for back-end (LDA-PLDA) training.”
Deep Speaker Feature Learning for Text-Independent Speaker Verification (2017, methodology): “, the SVM model for the GMM-UBM approach [4] and the PLDA model for the i-vector approach [5].”
Attentive Statistics Pooling for Deep Speaker Embedding (2018, methodology): “Mean subtraction, whitening, and length normalization [17] were applied to the i-vector as pre-processing steps before sending it to the PLDA, and similarity was then evaluated using a PLDA model with a speaker space of 400 dimensions.”
Bayesian Face Revisited: A Joint Formulation (2012, methodology): “LDA is based our own implementation and PLDA is from authors's implementation [19].”
In defence of metric learning for speaker recognition (2020, methodology): “Therefore, softmax-trained models have often been combined with PLDA [9] back-ends to generate scoring functions [5, 10].”
JHU Kaldi system for Arabic MGB-3 ASR challenge using diarization, audio-transcript alignment and transfer learning (2017, methodology): “However, in our experiments in English broadcast news, we found the actual PLDA log-likelihood ratio was more robust.”
ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification (2020, background): “In addition, a more complicated scoring backend can be trained such as Probabilistic Linear Discriminant Analysis (PLDA) [2].”
Acoustic Scene Classification: Classifying environments from the sounds they produce (2014, unknown): “a generative probabilistic linear discriminant analysis (PLDA) [30].”
Spot the conversation: speaker diarisation in the wild (2020, methodology): “Speaker embeddings are extracted using the x-vector [37] system, and the similarities between the embeddings are scored with a pre-trained probabilistic linear discriminant analysis (PLDA) [38, 39] model also provided in the code.”
X-Vectors: Robust DNN Embeddings for Speaker Recognition (2018, methodology): “A probabilistic linear discriminant analysis (PLDA) [3] classifier is used to compare i-vectors, and enable same-or-different speaker decisions [4, 5, 6].”
The Idlab Voxsrc-20 Submission: Large Margin Fine-Tuning and Quality-Aware Score Calibration in DNN Based Speaker Verification (2020, methodology): “An alternative Bayesian scoring technique is Probabilistic Linear Discriminant Analysis (PLDA) [8].”
AISHELL-4: An Open Source Dataset for Speech Enhancement, Separation, Recognition and Speaker Diarization in Conference Scenario (2021, methodology): “The post processing result of speaker embedding is derived from probabilistic linear discriminant analysis (PLDA) [28] model which is trained on the same data as the speaker embedding network.”
ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech (2019, methodology): “The original Kaldi recipe was modiﬁed to include PLDA adaptation using disjoint, bona ﬁde, in-domain data.”
Speaker Recognition for Multi-speaker Conversations Using X-vectors (2019, methodology): “A probabilistic linear discriminant analysis (PLDA) [3] classifier is used to compare i-vectors, and enable same-or-different speaker decisions [4, 5].”
Voxceleb: Large-scale speaker verification in the wild (2020, methodology): “Probabilistic LDA (PLDA) (Ioffe, 2006) is then used to reduce the dimension of the i-vectors to 200.”