# HyWA: Architecture-Preserving Personalized Voice Activity Detection for Full-Duplex Voice Assistants

arXiv: https://arxiv.org/abs/2510.12947 · HTML: https://mlsanity.com/paper/2510.12947/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 18 references that are in the map.
- First posted on arXiv 14 Oct 2025.

## Builds on (most cited first, 13 of 13)

- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [HyperNetworks](https://mlsanity.com/paper/1609.09106/) (2016)
- [MUSAN: A Music, Speech, and Noise Corpus](https://mlsanity.com/paper/1510.08484/) (2015)
- [Generalized End-to-End Loss for Speaker Verification](https://mlsanity.com/paper/1710.10467/) (2017)
- [FunASR: A Fundamental End-to-End Speech Recognition Toolkit](https://mlsanity.com/paper/2305.11013/) (2023)
- [VoiceFilter: Targeted Voice Separation by Speaker-Conditioned Spectrogram Masking](https://mlsanity.com/paper/1810.04826/) (2018)
- [Target-Speaker Voice Activity Detection: a Novel Approach for Multi-Speaker Diarization in a Dinner Party Scenario](https://mlsanity.com/paper/2005.07272/) (2020)
- [Personal VAD: Speaker-Conditioned Voice Activity Detection](https://mlsanity.com/paper/1908.04284/) (2019)
- [MarbleNet: Deep 1D Time-Channel Separable Convolutional Neural Network for Voice Activity Detection](https://mlsanity.com/paper/2010.13886/) (2020)
- [Personal VAD 2.0: Optimizing Personal Voice Activity Detection for On-Device Speech Recognition](https://mlsanity.com/paper/2204.03793/) (2022)
- [SVVAD: Personal Voice Activity Detection for Speaker Verification](https://mlsanity.com/paper/2305.19581/) (2023)
- [Comparative Analysis of Personalized Voice Activity Detection Systems: Assessing Real-World Effectiveness](https://mlsanity.com/paper/2406.09443/) (2024)
- [Tiny Noise-Robust Voice Activity Detector for Voice Assistants](https://mlsanity.com/paper/2507.22157/) (2025)

## Abstract (arXiv)

Voice activity detection (VAD) serves as an early gate in voice-assistant pipelines for smart devices. Because conventional VADs respond to speech from any speaker, nearby conversations and residual assistant playback lead to unwanted triggers, degrade the user experience, and waste computational resources. Personalized voice activity detection (PVAD) addresses this limitation by detecting speech only from an enrolled target speaker. Existing PVAD methods typically incorporate speaker information into model inputs or hidden representations. These approaches require VAD architectural changes that increase engineering and requalification costs in deployment. We present HyWA, a hypernetwork-based weight-adaptation method that converts an established VAD into a PVAD, while preserving its acoustic interface and inference topology. HyWA generates speaker-conditioned weights once at enrollment and requires no per-user optimization. We apply HyWA to pretrained Alibaba's FSMN and NVIDIA's MarbleNet VAD models. Evaluations on synthetic and real-user test data show consistent improvements in target-speaker discrimination and false-positive suppression. In an integrated full-duplex barge-in pipeline, replacing generic VAD with HyWA-based PVAD reduces observed false-interruption detections from 88.9% to only 9.9%.