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

This work presents a text-to-audio-retrieval system based on pre-trained textand spectrogram transformers. Our method projects recordings and textualdescriptions into a shared audio-caption space in which related examples fromdifferent modalities are close. Through a systematic analysis, we examine howeach component of the system influences retrieval performance. As a result, weidentify two key components that play a crucial role in driving performance:the self-attention-based audio encoder for audio embedding and the utilizationof additional human-generated and synthetic data sets during pre-training. Wefurther experimented with augmenting ClothoV2 captions with available keywordsto increase their variety; however, this only led to marginal improvements. Oursystem ranked first in the 2023's DCASE Challenge, and it outperforms thecurrent state of the art on the ClothoV2 benchmark by 5.6 pp. mAP@10.

Advancing Natural-Language Based Audio Retrieval with PaSST and Large Audio-Caption Data Sets