Picture for Themos Stafylakis

Themos Stafylakis

Challenging margin-based speaker embedding extractors by using the variational information bottleneck

Add code
Jun 18, 2024
Viaarxiv icon

Comparing Data Augmentation Methods for End-to-End Task-Oriented Dialog Systems

Add code
Jun 10, 2024
Viaarxiv icon

Do End-to-End Neural Diarization Attractors Need to Encode Speaker Characteristic Information?

Add code
Feb 29, 2024
Viaarxiv icon

DiaPer: End-to-End Neural Diarization with Perceiver-Based Attractors

Add code
Dec 22, 2023
Viaarxiv icon

A Simple Baseline for Knowledge-Based Visual Question Answering

Add code
Oct 24, 2023
Figure 1 for A Simple Baseline for Knowledge-Based Visual Question Answering
Figure 2 for A Simple Baseline for Knowledge-Based Visual Question Answering
Figure 3 for A Simple Baseline for Knowledge-Based Visual Question Answering
Figure 4 for A Simple Baseline for Knowledge-Based Visual Question Answering
Viaarxiv icon

Improving Speaker Verification with Self-Pretrained Transformer Models

May 17, 2023
Figure 1 for Improving Speaker Verification with Self-Pretrained Transformer Models
Figure 2 for Improving Speaker Verification with Self-Pretrained Transformer Models
Figure 3 for Improving Speaker Verification with Self-Pretrained Transformer Models
Figure 4 for Improving Speaker Verification with Self-Pretrained Transformer Models
Viaarxiv icon

Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing

Add code
Nov 03, 2022
Figure 1 for Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing
Figure 2 for Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing
Figure 3 for Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing
Figure 4 for Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing
Viaarxiv icon

Parameter-efficient transfer learning of pre-trained Transformer models for speaker verification using adapters

Oct 28, 2022
Figure 1 for Parameter-efficient transfer learning of pre-trained Transformer models for speaker verification using adapters
Figure 2 for Parameter-efficient transfer learning of pre-trained Transformer models for speaker verification using adapters
Figure 3 for Parameter-efficient transfer learning of pre-trained Transformer models for speaker verification using adapters
Figure 4 for Parameter-efficient transfer learning of pre-trained Transformer models for speaker verification using adapters
Viaarxiv icon

Extracting speaker and emotion information from self-supervised speech models via channel-wise correlations

Add code
Oct 15, 2022
Figure 1 for Extracting speaker and emotion information from self-supervised speech models via channel-wise correlations
Figure 2 for Extracting speaker and emotion information from self-supervised speech models via channel-wise correlations
Figure 3 for Extracting speaker and emotion information from self-supervised speech models via channel-wise correlations
Figure 4 for Extracting speaker and emotion information from self-supervised speech models via channel-wise correlations
Viaarxiv icon

On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding

Oct 11, 2022
Figure 1 for On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding
Figure 2 for On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding
Figure 3 for On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding
Figure 4 for On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding
Viaarxiv icon