EBEN: Extreme bandwidth extension network applied to speech signals captured with noise-resilient microphones
| Autor: | Hauret, Julien, Joubaud, Thomas, Zimpfer, Véronique, Bavu, Éric |
|---|---|
| Přispěvatelé: | BAVU, Eric, Intelligence artificielle pour la santé, la physqiue, les transports et la sécurité - - AHEAD2020 - ANR-20-THIA-0002 - PNIA - VALID |
| Jazyk: | angličtina |
| Rok vydání: | 2022 |
| Předmět: |
[SPI.ACOU] Engineering Sciences [physics]/Acoustics [physics.class-ph]
Deep Learning Bandwidth extension In-ear microphones Generative adversarial network Pseudo-quadrature filerbanks [STAT.ML] Statistics [stat]/Machine Learning [stat.ML] [SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing |
| Popis: | In this paper, we present Extreme Bandwidth Extension Network (EBEN), a generative adversarial network (GAN) that enhances audio measured with noise-resilient microphones. This type of capture equipment suppresses ambient noise at the expense of speech bandwidth, thereby requiring signal enhancement techniques to recover the wideband speech signal. EBEN leverages a multiband decomposition of the raw captured speech to decrease the data time-domain dimensions, and give better control over the full-band signal. This multiband representation is fed to a U-Net-like model, which adopts a combination of feature and adversarial losses to recover an enhanced audio signal. We also benefit from this original representation in the proposed discriminator architecture. Our approach can achieve state-of-the-art results with a lightweight generator and real-time compatible operation. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|