Experimenting with additive margins for contrastive self-supervised speaker verification
In Proc. Interspeech 2023
Abstract
A benchmark for toxic comment classification on civil comments dataset
In Extraction et gestion des connaissances, EGC 2023, lyon, france, 16 au 20 janvier 2023
Abstract
Label-efficient self-supervised speaker verification with information maximization and contrastive learning
In Proc. Interspeech 2022
Abstract
Hate speech and toxic comment detection using transformers
In Workshop EGC 2022 DL for NLP
Abstract Hate speech and toxic comment detection on social media has proven to be an essential issue for content moderation. This paper displays a comparison between different Transformer models for Hate Speech detection such as Hate BERT, a BERT-based model, RoBERTa and BERTweet which is a RoBERTa based model. These Transformer models are tested on Jibes&Delight 2021 reddit dataset using the same training and testing conditions. Multiple approaches are detailed in this paper considering feature extraction and data augmentation.
State-of-the-art speaker recognition with neural network embeddings in NIST SRE18 and speakers in the wild evaluations
In Computer Speech & Language
Abstract
State-of-the-art speaker recognition for telephone and video speech: The JHU-MIT submission for NIST SRE18
In Proc. Interspeech 2019
Abstract
The MIT Lincoln Laboratory 2016 speaker recognition system
In NIST speaker recognition evaluation 2016
Abstract This document presents the system submission for the group composed of MIT Lincoln Laboratory, Johns Hopkins University (JHU), Laboratoire de Recherche et de Développement de l’EPITA (LRDE) and Universidad Autónoma de Madrid (ATVS). The primary submission is a combination of four systems focused on i-vector systems. Two secondary submissions are also included
GMM weights adaptation based on subspace approaches for speaker verification
In Odyssey 2014, the speaker and language recognition workshop
Abstract In this paper, we explored the use of Gaussian Mixture Model (GMM) weights adaptation for speaker verifica- tion. We compared two different subspace weight adap- tation approaches: Subspace Multinomial Model (SMM) and Non-Negative factor Analysis (NFA). Both techniques achieved similar results and seemed to outperform the retraining maximum likelihood (ML) weight adaptation. However, the training process for the NFA approach is substantially faster than the SMM technique. The i-vector fusion between each weight adaptation approach and the classical i-vector yielded slight improvements on the tele- phone part of the NIST 2010 Speaker Recognition Eval- uation dataset.
The MIT lincoln laboratory / JHU / EPITA-LSE LRE17 system
In Odyssey 2018, the speaker and language recognition workshop
Abstract
Unsupervised methods for speaker diarization: An integrated and iterative approach
In IEEE Transactions on Audio, Speech, and Language Processing
Abstract In speaker diarization, standard approaches typically perform speaker clustering on some initial segmentation before refining the segment boundaries in a re-segmentation step to obtain a final diarization hypothesis. In this paper, we integrate an improved clustering method with an existing re-segmentation algorithm and, in iterative fashion, optimize both speaker cluster assignments and segmentation boundaries jointly. For clustering, we extend our previous research using factor analysis for speaker modeling. In continuing to take advantage of the effectiveness of factor analysis as a front-end for extracting speaker-specific features (i.