# Streaming Endpointer for Spoken Dialogue using Neural Audio Codecs and Label-Delayed Training

arXiv: https://arxiv.org/abs/2506.07081 · HTML: https://mlsanity.com/paper/2506.07081/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Jun 2025.

## Built on by (most cited first, 4 of 4)

- [KAME: Tandem Architecture for Enhancing Knowledge in Real-Time Speech-to-Speech Conversational AI](https://mlsanity.com/paper/2510.02327/) (2025)
- [TurnBench: A Multi-Domain Benchmark for Turn-Taking Dynamics in Spoken Dialogue](https://mlsanity.com/paper/2608.25218/) (2026)
- [DualTurn: Learning Turn-Taking from Dual-Channel Generative Speech Pretraining](https://mlsanity.com/paper/2603.08216/) (2026)
- [MultiPartyTurnBench: Benchmarking Floor Control in Multi-Party Spoken Dialogue](https://mlsanity.com/paper/2610.20172/) (2026)

## Abstract (arXiv)

Accurate, low-latency endpointing is crucial for effective spoken dialogue systems. While traditional endpointers often rely on spectrum-based audio features, this work proposes real-time speech endpointing for multi-turn dialogues using streaming, low-bitrate Neural Audio Codec (NAC) features, building upon recent advancements in neural audio codecs. To further reduce cutoff errors, we introduce a novel label delay training scheme. At a fixed median latency of 160 ms, our combined NAC and label delay approach achieves significant relative cutoff error reductions: 42.7% for a single-stream endpointer and 37.5% for a two-stream configuration, compared to baseline methods. Finally, we demonstrate efficient integration with a codec-based pretrained speech large language model, improving its median response time by 1200 ms and reducing its cutoff error by 35%.