{"slug": "beyond-wer-entity-and-disfluency-recall-in-accented-conversational-asr", "title": "Beyond WER: Entity and Disfluency Recall in Accented Conversational ASR", "summary": "A three-stage ASR pipeline built on Qwen2.5-Omni-3B achieved 80-85% entity recall and 76-86% filler recall across 6,000 test utterances from speakers in India, Indonesia, and Latin America, up from 53-55% and under 5% respectively, according to the arXiv paper 2609.20828v1. The pipeline combines heuristic SQL filters that curate entity-rich training data at 2.8x the entity density of random sampling, regional LoRA adapters that emit verbatim and corrected transcripts in a single forward pass, and a six-category error taxonomy validated by an LLM-based judge at 83.8% agreement on 210 human-labelled samples. It reached 6-10% WER, outperformed Whisper and a commercial ASR on entity recall, and matched a zero-shot 30B model with 10x fewer parameters, with paired bootstrap tests attributing 2.8-4.2 percentage points of entity recall gain to curation alone (p<0.0001).", "body_md": "arXiv:2609.20828v1 Announce Type: new \nAbstract: ASR systems optimised for Word Error Rate (WER) often miss named entities and filled pauses in accented conversational English, both critical for language-learning feedback. We present a three-stage pipeline for speakers from India, Indonesia, and Latin America: (1) heuristic SQL filters curating entity-rich training data at 2.8x the entity density of random sampling, (2) regional LoRA adapters fine-tuned on Qwen2.5-Omni-3B producing both verbatim and corrected transcripts in a single forward pass, and (3) a six-category error taxonomy validated by an LLM-based judge (83.8% agreement, 210 human-labelled samples). The pipeline achieves 80-85% entity recall (up from 53-55%), 76-86% filler recall (up from <5%), and 6-10% WER across 6k test utterances, outperforming Whisper and a commercial ASR on entity recall while matching a zero-shot 30B model with 10x fewer parameters. Paired bootstrap tests confirm that curation alone accounts for 2.8-4.2 pp of entity recall gain (p<0.0001).", "url": "https://wpnews.pro/news/beyond-wer-entity-and-disfluency-recall-in-accented-conversational-asr", "canonical_source": "https://arxiv.org/abs/2609.20828", "published_at": "2026-09-21 04:00:00+00:00", "updated_at": "2026-09-21 04:23:26.851121+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "natural-language-processing", "ai-research", "large-language-models"], "entities": ["Qwen2.5-Omni-3B", "Whisper", "arXiv", "India", "Indonesia", "Latin America"], "alternates": {"html": "https://wpnews.pro/news/beyond-wer-entity-and-disfluency-recall-in-accented-conversational-asr", "markdown": "https://wpnews.pro/news/beyond-wer-entity-and-disfluency-recall-in-accented-conversational-asr.md", "text": "https://wpnews.pro/news/beyond-wer-entity-and-disfluency-recall-in-accented-conversational-asr.txt", "jsonld": "https://wpnews.pro/news/beyond-wer-entity-and-disfluency-recall-in-accented-conversational-asr.jsonld"}}