Picture for Georgios Paraskevopoulos

Georgios Paraskevopoulos

Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens

Add code
Feb 03, 2024
Figure 1 for Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens
Figure 2 for Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens
Figure 3 for Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens
Figure 4 for Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens
Viaarxiv icon

Investigating Personalization Methods in Text to Music Generation

Add code
Sep 20, 2023
Figure 1 for Investigating Personalization Methods in Text to Music Generation
Figure 2 for Investigating Personalization Methods in Text to Music Generation
Figure 3 for Investigating Personalization Methods in Text to Music Generation
Figure 4 for Investigating Personalization Methods in Text to Music Generation
Viaarxiv icon

Weakly-supervised forced alignment of disfluent speech using phoneme-level modeling

Add code
May 30, 2023
Figure 1 for Weakly-supervised forced alignment of disfluent speech using phoneme-level modeling
Figure 2 for Weakly-supervised forced alignment of disfluent speech using phoneme-level modeling
Figure 3 for Weakly-supervised forced alignment of disfluent speech using phoneme-level modeling
Figure 4 for Weakly-supervised forced alignment of disfluent speech using phoneme-level modeling
Viaarxiv icon

Depression detection in social media posts using affective and social norm features

Add code
Mar 24, 2023
Figure 1 for Depression detection in social media posts using affective and social norm features
Figure 2 for Depression detection in social media posts using affective and social norm features
Figure 3 for Depression detection in social media posts using affective and social norm features
Figure 4 for Depression detection in social media posts using affective and social norm features
Viaarxiv icon

Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems A case study for Modern Greek

Add code
Dec 31, 2022
Figure 1 for Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems A case study for Modern Greek
Figure 2 for Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems A case study for Modern Greek
Figure 3 for Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems A case study for Modern Greek
Figure 4 for Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems A case study for Modern Greek
Viaarxiv icon

Adapted Multimodal BERT with Layer-wise Fusion for Sentiment Analysis

Dec 01, 2022
Figure 1 for Adapted Multimodal BERT with Layer-wise Fusion for Sentiment Analysis
Figure 2 for Adapted Multimodal BERT with Layer-wise Fusion for Sentiment Analysis
Figure 3 for Adapted Multimodal BERT with Layer-wise Fusion for Sentiment Analysis
Figure 4 for Adapted Multimodal BERT with Layer-wise Fusion for Sentiment Analysis
Viaarxiv icon

Extending Compositional Attention Networks for Social Reasoning in Videos

Add code
Oct 03, 2022
Figure 1 for Extending Compositional Attention Networks for Social Reasoning in Videos
Figure 2 for Extending Compositional Attention Networks for Social Reasoning in Videos
Figure 3 for Extending Compositional Attention Networks for Social Reasoning in Videos
Figure 4 for Extending Compositional Attention Networks for Social Reasoning in Videos
Viaarxiv icon

Regotron: Regularizing the Tacotron2 architecture via monotonic alignment loss

Add code
Apr 28, 2022
Figure 1 for Regotron: Regularizing the Tacotron2 architecture via monotonic alignment loss
Figure 2 for Regotron: Regularizing the Tacotron2 architecture via monotonic alignment loss
Figure 3 for Regotron: Regularizing the Tacotron2 architecture via monotonic alignment loss
Figure 4 for Regotron: Regularizing the Tacotron2 architecture via monotonic alignment loss
Viaarxiv icon

MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis

Add code
Jan 24, 2022
Figure 1 for MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis
Figure 2 for MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis
Figure 3 for MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis
Figure 4 for MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis
Viaarxiv icon

EmpBot: A T5-based Empathetic Chatbot focusing on Sentiments

Oct 30, 2021
Figure 1 for EmpBot: A T5-based Empathetic Chatbot focusing on Sentiments
Figure 2 for EmpBot: A T5-based Empathetic Chatbot focusing on Sentiments
Figure 3 for EmpBot: A T5-based Empathetic Chatbot focusing on Sentiments
Figure 4 for EmpBot: A T5-based Empathetic Chatbot focusing on Sentiments
Viaarxiv icon