79 citations · 91 across the 18 of their papers we have counts for
23 papers
WeDefense: A Toolkit to Defend Against Fake Audio
Lin Zhang, Johan Rohdin, Xin Wang +8
The advances in generative AI have enabled the creation of synthetic audio which is perceptually indistinguishable from real, genuine audio. Although this stellar progress enables…
Fine-tune Before Structured Pruning: Towards Compact and Accurate Self-Supervised Models for Speaker Diarization
Jiangyu Han, Federico Landini, Johan Rohdin +4
Self-supervised learning (SSL) models like WavLM can be effectively utilized when building speaker diarization systems but are often large and slow, limiting their use in resource…
Analysis of ABC Frontend Audio Systems for the NIST-SRE24
Sara Barahona, Anna Silnova, Ladislav Mošner +14
We present a comprehensive analysis of the embedding extractors (frontends) developed by the ABC team for the audio track of NIST SRE 2024. We follow the two scenarios imposed by N…
Joint Training of Speaker Embedding Extractor, Speech and Overlap Detection for Diarization
Petr Pálka, Federico Landini, Dominik Klement +4
In spite of the popularity of end-to-end diarization systems nowadays, modular systems comprised of voice activity detection (VAD), speaker embedding extraction plus clustering, an…
Leveraging Self-Supervised Learning for Speaker Diarization
Jiangyu Han, Federico Landini, Johan Rohdin +3
End-to-end neural diarization has evolved considerably over the past few years, but data scarcity is still a major obstacle for further improvements. Self-supervised learning metho…
BUT Systems and Analyses for the ASVspoof 5 Challenge
Johan Rohdin, Lin Zhang, Oldřich Plchot +8
This paper describes the BUT submitted systems for the ASVspoof 5 challenge, along with analyses. For the conventional deepfake detection task, we use ResNet18 and self-supervised…
Challenging margin-based speaker embedding extractors by using the variational information bottleneck
Themos Stafylakis, Anna Silnova, Johan Rohdin +2
Speaker embedding extractors are typically trained using a classification loss over the training speakers. During the last few years, the standard softmax/cross-entropy loss has be…
Do End-to-End Neural Diarization Attractors Need to Encode Speaker Characteristic Information?
Lin Zhang, Themos Stafylakis, Federico Landini +3
In this paper, we apply the variational information bottleneck approach to end-to-end neural diarization with encoder-decoder attractors (EEND-EDA). This allows us to investigate w…
Discriminative Training of VBx Diarization
Dominik Klement, Mireia Diez, Federico Landini +4
Bayesian HMM clustering of x-vector sequences (VBx) has become a widely adopted diarization baseline model in publications and challenges. It uses an HMM to model speaker turns, a…
Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx) for Combined End-to-End and Vector Clustering-based Diarization
Marc Delcroix, Naohiro Tawara, Mireia Diez +6
Combining end-to-end neural speaker diarization (EEND) with vector clustering (VC), known as EEND-VC, has gained interest for leveraging the strengths of both methods. EEND-VC esti…
Toroidal Probabilistic Spherical Discriminant Analysis
Anna Silnova, Niko Brümmer, Albert Swart +1
In speaker recognition, where speech segments are mapped to embeddings on the unit hypersphere, two scoring back-ends are commonly used, namely cosine scoring and PLDA. We have rec…
Probabilistic Spherical Discriminant Analysis: An Alternative to PLDA for length-normalized embeddings
Niko Brümmer, Albert Swart, Ladislav Mošner +4
In speaker recognition, where speech segments are mapped to embeddings on the unit hypersphere, two scoring backends are commonly used, namely cosine scoring or PLDA. Both have adv…
Analyzing speaker verification embedding extractors and back-ends under language and channel mismatch
Anna Silnova, Themos Stafylakis, Ladislav Mosner +6
In this paper, we analyze the behavior and performance of speaker embeddings and the back-end scoring model under domain and language mismatch. We present our findings regarding Re…
Training Speaker Embedding Extractors Using Multi-Speaker Audio with Unknown Speaker Boundaries
Themos Stafylakis, Ladislav Mošner, Oldřich Plchot +4
In this paper, we demonstrate a method for training speaker embedding extractors using weak annotation. More specifically, we are using the full VoxCeleb recordings and the name of…
Analysis of the BUT Diarization System for VoxConverse Challenge
Federico Landini, Ondřej Glembek, Pavel Matějka +4
This paper describes the system developed by the BUT team for the fourth track of the VoxCeleb Speaker Recognition Challenge, focusing on diarization on the VoxConverse dataset. Th…
Probabilistic embeddings for speaker diarization
Anna Silnova, Niko Brümmer, Johan Rohdin +2
Speaker embeddings (x-vectors) extracted from very short segments of speech have recently been shown to give competitive performance in speaker diarization. We generalize this reci…
BUT System for the Second DIHARD Speech Diarization Challenge
Federico Landini, Shuai Wang, Mireia Diez +9
This paper describes the winning systems developed by the BUT team for the four tracks of the Second DIHARD Speech Diarization Challenge. For tracks 1 and 2 the systems were mainly…
BUT System Description to VoxCeleb Speaker Recognition Challenge 2019
Hossein Zeinali, Shuai Wang, Anna Silnova +2
In this report, we describe the submission of Brno University of Technology (BUT) team to the VoxCeleb Speaker Recognition Challenge (VoxSRC) 2019. We also provide a brief analysis…
BUT VOiCES 2019 System Description
Hossein Zeinali, Pavel Matějka, Ladislav Mošner +6
This is a description of our effort in VOiCES 2019 Speaker Recognition challenge. All systems in the fixed condition are based on the x-vector paradigm with different features and…
Speaker verification using end-to-end adversarial language adaptation
Johan Rohdin, Themos Stafylakis, Anna Silnova +3
In this paper we investigate the use of adversarial domain adaptation for addressing the problem of language mismatch between speaker recognition corpora. In the context of speaker…
Fast variational Bayes for heavy-tailed PLDA applied to i-vectors and x-vectors
Anna Silnova, Niko Brummer, Daniel Garcia-Romero +2
The standard state-of-the-art backend for text-independent speaker recognizers that use i-vectors or x-vectors, is Gaussian PLDA (G-PLDA), assisted by a Gaussianization step involv…
Gaussian meta-embeddings for efficient scoring of a heavy-tailed PLDA model
Niko Brummer, Anna Silnova, Lukas Burget +1
Embeddings in machine learning are low-dimensional representations of complex input patterns, with the property that simple geometric operations like Euclidean distances and dot pr…
End-to-end DNN Based Speaker Recognition Inspired by i-vector and PLDA
Johan Rohdin, Anna Silnova, Mireia Diez +3
Recently several end-to-end speaker verification systems based on deep neural networks (DNNs) have been proposed. These systems have been proven to be competitive for text-dependen…