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

Previous studies in automated audio captioning have faced difficulties inaccurately capturing the complete temporal details of acoustic scenes andevents within long audio sequences. This paper presents AudioLog, a largelanguage models (LLMs)-powered audio logging system with hybrid token-semanticcontrastive learning. Specifically, we propose to fine-tune the pre-trainedhierarchical token-semantic audio Transformer by incorporating contrastivelearning between hybrid acoustic representations. We then leverage LLMs togenerate audio logs that summarize textual descriptions of the acousticenvironment. Finally, we evaluate the AudioLog system on two datasets with bothscene and event annotations. Experiments show that the proposed system achievesexceptional performance in acoustic scene classification and sound eventdetection, surpassing existing methods in the field. Further analysis of theprompts to LLMs demonstrates that AudioLog can effectively summarize long audiosequences. To the best of our knowledge, this approach is the first attempt toleverage LLMs for summarizing long audio sequences.

AudioLog: LLMs-Powered Long Audio Logging with Hybrid Token-Semantic Contrastive Learning