# Researchers test pseudo‑labeling to improve ASR on noisy police audio

Digest AI · Generative AI & Models · published 2026-09-28T04:00:00Z

Canonical: https://digestai.news/story/researchers-test-pseudolabeling-to-improve-asr-on-noisy-police-audio

## Summary

Researchers evaluated pseudo‑labeling to adapt large ASR models to noisy police communication data from Baltimore and Chicago. They found that standard confidence metrics such as log‑probabilities and STAR scores cannot reliably separate high‑from low‑quality pseudo‑labels.

To improve filtering, the team introduced an external LLM‑as‑judge approach that discards contextually implausible transcripts. This method reduced word‑error‑rate (WER) more than internal metrics, though a gap remains compared to an oracle filter. They also explored a cross‑model pseudo‑labeling strategy, where one model is fine‑tuned with pseudo‑labels generated by another, and identified it as a promising direction for future work.

## Key points

- Pseudo‑labeling improves ASR on noisy police audio but internal confidence metrics fail to distinguish quality
- External LLM‑as‑judge filtering reduces WER more than internal metrics
- Cross‑model pseudo‑labeling shows promise for future ASR adaptation

## Why it matters

Improving ASR on noisy police audio can aid law‑enforcement analysis and public transparency, while highlighting limits of current confidence metrics in noisy domains.

## Sources

1. [Pretrained ASR Pseudo-labeling for Noisy Police Audio](https://arxiv.org/abs/2609.30469) (arXiv cs.AI, 2026-09-28, primary source)

Part of the developing story: [AI Whisper Advances Multilingual Speech Recognition](https://digestai.news/thread/token-merging-boosts-whisper-efficiency-across-16-languages-with-minimal) (3 stories)

## Cite

Digest AI, "Researchers test pseudo‑labeling to improve ASR on noisy police audio", 28 September 2026, https://digestai.news/story/researchers-test-pseudolabeling-to-improve-asr-on-noisy-police-audio

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/researchers-test-pseudolabeling-to-improve-asr-on-noisy-police-audio.json