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

Humans possess an extraordinary ability to selectively focus on the soundsource of interest amidst complex acoustic environments, commonly referred toas cocktail party scenarios. In an attempt to replicate this remarkableauditory attention capability in machines, target speaker extraction (TSE)models have been developed. These models leverage the pre-registered cues ofthe target speaker to extract the sound source of interest. However, theeffectiveness of these models is hindered in real-world scenarios due to theunreliable or even absence of pre-registered cues. To address this limitation,this study investigates the integration of natural language description toenhance the feasibility, controllability, and performance of existing TSEmodels. Specifically, we propose a model named LLM-TSE, wherein a largelanguage model (LLM) extracts useful semantic cues from the user's typed textinput. These cues can serve as independent extraction cues, task selectors tocontrol the TSE process or complement the pre-registered cues. Our experimentalresults demonstrate competitive performance when only text-based cues arepresented, the effectiveness of using input text as a task selector, and a newstate-of-the-art when combining text-based cues with pre-registered cues. Toour knowledge, this is the first study to successfully incorporate LLMs toguide target speaker extraction, which can be a cornerstone for cocktail partyproblem research.

Typing to Listen at the Cocktail Party: Text-Guided Target Speaker Extraction