4 papers
Efficient Audio Enhancement with a Differentiable Psychoacoustic Loss
Wallace Abreu, Bernardo V. Miranda, Luiz W. P. Biscainho
Audio enhancement consists of improving the perceived quality of audio signals. Initially, with the aim of addressing bandwidth extension, this work proposes \(AEROMamba_{P}\), an…
Extracting accent features in spoken Brazilian Portuguese without sociolinguistic labels
Pedro H. L. Leite, Pedro Benevenuto Valadares, Luiz W. P. Biscainho
Regional accent classification in Brazilian Portuguese (pt-BR) suffers from the need for reliable labeling. While large self-supervised learning (SSL) speech models are powerful, t…
FiPA-SR -- FiLM-Conditioned Perceptually Informed Audio Super-Resolution
Wallace Abreu, Luiz W. P. Biscainho
Audio bandwidth extension aims to reconstruct missing high-frequency content from bandlimited signals. This paper proposes FiPA-SR, a GAN-based perceptual architecture capable of h…
AEROMamba: An efficient architecture for audio super-resolution using generative adversarial networks and state space models
Wallace Abreu, Luiz Wagner Pereira Biscainho
Audio super-resolution aims to enhance low-resolution signals by creating high-frequency content. In this work, we modify the architecture of AERO (a state-of-the-art system for th…