English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The synthesis of sound via deep learning methods has recently received muchattention. Some problems for deep learning approaches to sound synthesis relateto the amount of data needed to specify an audio signal and the necessity ofpreserving both the long and short time coherence of the synthesised signal.Visual time-frequency representations such as the log-mel-spectrogram havegained in popularity. The log-mel-spectrogram is a perceptually informedrepresentation of audio that greatly compresses the amount of informationrequired for the description of the sound. However, because of thiscompression, this representation is not directly invertible. Both signalprocessing and machine learning techniques have previously been applied to theinversion of the log-mel-spectrogram but they both caused audible distortionsin the synthesized sounds due to issues of temporal and spectral coherence. Inthis paper, we outline the application of a sinusoidal model to the inversionof the log-mel-spectrogram for pitched musical instrument sounds outperformingstate-of-the-art deep learning methods. The approach could be later used as ageneral decoding step from spectral to time intervals in neural applications.

A sinusoidal signal reconstruction method for the inversion of the mel-spectrogram