| title | VoxLock: Target-Speaker Extraction for Voice Agents Without an Enrollment Step |
|---|---|
| authors | Arbaz Siddiqui |
| date | 2026-10-29 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.40727 |
| upvotes | 3 (Hugging Face Daily Papers, 2026-11-03) |
| links | abs pdf hf code |
| abstract | A voice agent's speech recognizer transcribes everyone near the microphone, including people other than the caller. I fine-tuned a public causal target-speaker extraction model (REAL-TSE, 27M parameters) on one GPU to work from a voice sample cut from the call itself, and released it as VoxLock. I score it by what a recognizer writes on real recordings. On Krisp's 265-recording benchmark it reaches 4.95% WER against 23.21% unprocessed and lets 9 background words into the transcript over 85 minutes of background-only speech. Its WER is not significantly different from Krisp's best filters and is significantly lower than the other commercial and open systems I tested. On 400 AMI meeting windows built and frozen before scoring, it cuts WER from 42.95% to 10.63%. The 8-bit model runs a live call on 77% of one CPU thread with 24 ms of algorithmic latency. Model, code and per-file scores are open. |
cite
@misc{siddiqui2026voxlock,
title={VoxLock: Target-Speaker Extraction for Voice Agents Without an Enrollment Step},
author={Arbaz Siddiqui},
year={2026},
eprint={2610.40727},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.40727}
}