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

In audio processing applications, the generation of expressive sounds basedon high-level representations demonstrates a high demand. These representationscan be used to manipulate the timbre and influence the synthesis of creativeinstrumental notes. Modern algorithms, such as neural networks, have inspiredthe development of expressive synthesizers based on musical instrument timbrecompression. Unsupervised deep learning methods can achieve audio compressionby training the network to learn a mapping from waveforms or spectrograms tolow-dimensional representations. This study investigates the use of stackedconvolutional autoencoders for the compression of time-frequency audiorepresentations for a variety of instruments for a single pitch. Furtherexploration of hyper-parameters and regularization techniques is demonstratedto enhance the performance of the initial design. In an unsupervised manner,the network is able to reconstruct a monophonic and harmonic sound based onlatent representations. In addition, we introduce an evaluation metric tomeasure the similarity between the original and reconstructed samples.Evaluating a deep generative model for the synthesis of sound is a challengingtask. Our approach is based on the accuracy of the generated frequencies as itpresents a significant metric for the perception of harmonic sounds. This workis expected to accelerate future experiments on audio compression using neuralautoencoders.

An investigation of the reconstruction capacity of stacked convolutional autoencoders for log-mel-spectrograms