3 papers
eess.SP2026
Listen to the Features: Voice Anonymization Driven by Content Embedding Matching over Signal Reconstruction
Adrien Schneider, Kacper Zabkowski, Anderson Augusma +3
The paper presents a voice anonymization model focusing on preserving content rather than producing realistic speech. It relies on content embeddings extracted from a frozen pretra…
cs.CV2024
Exploring VQ-VAE with Prosody Parameters for Speaker Anonymization
Sotheara Leang, Anderson Augusma, Eric Castelli +3
Human speech conveys prosody, linguistic content, and speaker identity. This article investigates a novel speaker anonymization approach using an end-to-end network based on a Vect…
cs.AI2023
Multimodal Group Emotion Recognition In-the-wild Using Privacy-Compliant Features
Anderson Augusma, Dominique Vaufreydaz, Frédérique Letué
This paper explores privacy-compliant group-level emotion recognition ''in-the-wild'' within the EmotiW Challenge 2023. Group-level emotion recognition can be useful in many fields…