# The Speech-LLM Takes It All: A Truly Fully End-to-End Spoken Dialogue State Tracking Approach

Source: [The Speech-LLM Takes It All: A Truly Fully End-to-End Spoken Dialogue State Tracking Approach](<https://www.sota2.com/research/paper/68fa7c0d1e209131cea8c22b>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-18T05:00:15.522Z

Authors: Nizar El Ghazal, Antoine Caubri\\\`ere, Valentin Vielzeuf

Published: 2025-10-10T14:27:01.000Z

## Abstract

This paper presents a comparative study of context management strategies for end-to-end Spoken Dialog State Tracking using Speech-LLMs. We systematically evaluate traditional multimodal context (combining text history and spoken current turn), full spoken history, and compressed spoken history approaches. Our experiments on the SpokenWOZ corpus demonstrate that providing the full spoken conversation as input yields the highest performance among models of similar size, significantly surpassing prior methods. Furthermore, we show that attention-pooling-based compression of the spoken history offers a strong trade-off, maintaining competitive accuracy with reduced context size. Detailed analysis confirms that improvements stem from more effective context utilization.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Dialog State Tracking on SpokenWoz (test)](<https://www.sota2.com/research/sota/dialog-state-tracking-on-spokenwoz-test.md>): Dialog State Tracking; SpokenWoz (test); JGA: 45.52.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2510.09424>)

[PDF](<https://arxiv.org/pdf/2510.09424.pdf>)