Guardado en:
Detalles Bibliográficos
Autores principales: Helwani, Karim, Do, Hoang, Luan, James, Srinivasan, Sriram
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2603.13379
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910052566171648
author Helwani, Karim
Do, Hoang
Luan, James
Srinivasan, Sriram
author_facet Helwani, Karim
Do, Hoang
Luan, James
Srinivasan, Sriram
contents We present a real-time front-end for voice-based conversational AI to enable natural turn-taking in two-speaker scenarios by combining primary speaker segmentation with hierarchical End-of-Turn (EOT) detection. To operate robustly in multi-speaker environments, the system continuously identifies and tracks the primary user, ensuring that downstream EOT decisions are not confounded by background conversations. The tracked activity segments are fed to a hierarchical, causal EOT model that predicts the immediate conversational state by independently analyzing per-speaker speech features from both the primary speaker and the bot. Simultaneously, the model anticipates near-future states ($t{+}10/20/30$\,ms) through probabilistic predictions that are aware of the conversation partner's speech. Task-specific knowledge distillation compresses wav2vec~2.0 representations (768\,D) into a compact MFCC-based student (32\,D) for efficient deployment. The system achieves 82\% multi-class frame-level F1 and 70.6\% F1 on Backchannel detection, with 69.3\% F1 on a binary Final vs.\ Others task. On an end-to-end turn-detection benchmark, our model reaches 87.7\% recall vs.\ 58.9\% for Smart Turn~v3 while keeping a median detection latency of 36\,ms versus 800--1300\,ms. Despite using only 1.14\,M parameters, the proposed model matches or exceeds transformer-based baselines while substantially reducing latency and memory footprint, making it suitable for edge deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2603_13379
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Hierarchical End-of-Turn Model with Primary Speaker Segmentation for Real-Time Conversational AI
Helwani, Karim
Do, Hoang
Luan, James
Srinivasan, Sriram
Machine Learning
Sound
We present a real-time front-end for voice-based conversational AI to enable natural turn-taking in two-speaker scenarios by combining primary speaker segmentation with hierarchical End-of-Turn (EOT) detection. To operate robustly in multi-speaker environments, the system continuously identifies and tracks the primary user, ensuring that downstream EOT decisions are not confounded by background conversations. The tracked activity segments are fed to a hierarchical, causal EOT model that predicts the immediate conversational state by independently analyzing per-speaker speech features from both the primary speaker and the bot. Simultaneously, the model anticipates near-future states ($t{+}10/20/30$\,ms) through probabilistic predictions that are aware of the conversation partner's speech. Task-specific knowledge distillation compresses wav2vec~2.0 representations (768\,D) into a compact MFCC-based student (32\,D) for efficient deployment. The system achieves 82\% multi-class frame-level F1 and 70.6\% F1 on Backchannel detection, with 69.3\% F1 on a binary Final vs.\ Others task. On an end-to-end turn-detection benchmark, our model reaches 87.7\% recall vs.\ 58.9\% for Smart Turn~v3 while keeping a median detection latency of 36\,ms versus 800--1300\,ms. Despite using only 1.14\,M parameters, the proposed model matches or exceeds transformer-based baselines while substantially reducing latency and memory footprint, making it suitable for edge deployment.
title A Hierarchical End-of-Turn Model with Primary Speaker Segmentation for Real-Time Conversational AI
topic Machine Learning
Sound
url https://arxiv.org/abs/2603.13379