English

Spoken Conversational Agents with Large Language Models

Computation and Language 2026-03-10 v1 Multiagent Systems Neural and Evolutionary Computing Sound Audio and Speech Processing

Abstract

Spoken conversational agents are converging toward voice-native LLMs. This tutorial distills the path from cascaded ASR/NLU to end-to-end, retrieval-and vision-grounded systems. We frame adaptation of text LLMs to audio, cross-modal alignment, and joint speech-text training; review datasets, metrics, and robustness across accents and compare design choices (cascaded vs. E2E, post-ASR correction, streaming). We link industrial assistants to current open-domain and task-oriented agents, highlight reproducible baselines, and outline open problems in privacy, safety, and evaluation. Attendees leave with practical recipes and a clear systems-level roadmap.

Keywords

Cite

@article{arxiv.2512.02593,
  title  = {Spoken Conversational Agents with Large Language Models},
  author = {Chao-Han Huck Yang and Andreas Stolcke and Larry Heck},
  journal= {arXiv preprint arXiv:2512.02593},
  year   = {2026}
}

Comments

Accepted to EMNLP 2025 Tutorial

R2 v1 2026-07-01T08:05:24.697Z