Hypernetworks build Implicit Neural Representations of Sounds
This provides a novel alternative to spectrograms for audio processing in deep neural networks, addressing a domain-specific gap in INR applications.
The authors tackled the challenge of applying Implicit Neural Representations (INRs) to audio data, which is nontrivial due to biases in image-based models, by introducing HyperSound, a meta-learning approach using hypernetworks that achieves audio reconstruction quality comparable to state-of-the-art models.
Implicit Neural Representations (INRs) are nowadays used to represent multimedia signals across various real-life applications, including image super-resolution, image compression, or 3D rendering. Existing methods that leverage INRs are predominantly focused on visual data, as their application to other modalities, such as audio, is nontrivial due to the inductive biases present in architectural attributes of image-based INR models. To address this limitation, we introduce HyperSound, the first meta-learning approach to produce INRs for audio samples that leverages hypernetworks to generalize beyond samples observed in training. Our approach reconstructs audio samples with quality comparable to other state-of-the-art models and provides a viable alternative to contemporary sound representations used in deep neural networks for audio processing, such as spectrograms.