AERO: Audio Super Resolution in the Spectral Domain

• Main Authors: Mandel, Moshe, Tal, Or, Adi, Yossi
• Format: Conference Proceeding
• Language: English
• Subjects: audio super-resolution; Bandwidth; bandwidth extension; Codes; Frequency-domain analysis; High frequency; Multiple signal classification; Speech processing; speech synthesis; Superresolution
• ISSN: 2379-190X
• DOI: 10.1109/ICASSP49357.2023.10095382

We present AERO, a audio super-resolution model that processes speech and music signals in the spectral domain. AERO is based on an encoder-decoder architecture with UNet like skip connections. We optimize the model using both time and frequency domain loss functions. Specifically, we consider a set of reconstruction losses together with perceptual ones in the form of adversarial and feature discriminator loss functions. To better handle phase information the proposed method operates over the complex-valued spectrogram using two separate channels. Unlike prior work which mainly considers low and high frequency concatenation for audio super-resolution, the proposed method directly predicts the full frequency range. We demonstrate high performance across a wide range of sample rates considering both speech and music. AERO outperforms the evaluated baselines considering Log-Spectral Distance, ViSQOL, and the subjective MUSHRA test. Audio samples and code are available {\color{Blue}\text{here}}.