---
title: "MedAstra: ASR-less Speech Retrieval with Diagnostic Grounding for Medical QA"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.01261v1"
permalink: "https://feed.astralor.com/items/itm_f51106eff75111e5b9d75504"
published: "2026-10-31T21:00:22.000Z"
score: 57
channel: "tech"
tags: ["RAG","MedAstra","ASR-less Speech Retrieval","Diagnostic Grounding","Medical","Spoken","ASR","WavLM"]
---

# MedAstra: ASR-less Speech Retrieval with Diagnostic Grounding for Medical QA

## 速览

Spoken medical question answering can fail when automatic speech recognition (ASR) corrupts a clinically important term before retrieval. We present MedAstra, a framework that maps frozen WavLM speech representations th…

## 关键点

- Spoken medical question answering can fail when automatic speech recognition (ASR) corrupts a clinically important term before retrieval.
- We present MedAstra, a framework that maps frozen WavLM speech representations through a lightweight adapter into a frozen E5 text embedding space.
- This enables a dense retrieval path that accepts a waveform without a query transcript.

## 全文

Spoken medical question answering can fail when automatic speech recognition (ASR) corrupts a clinically important term before retrieval. We present MedAstra, a framework that maps frozen WavLM speech representations through a lightweight adapter into a frozen E5 text embedding space. This enables a dense retrieval path that accepts a waveform without a query transcript. Retrieved passages and speech-linked textual metadata support answer generation with entity alternatives, diagnostic contrasts, relation heuristics, and bounded retrieval refinement. ASR-less refers to the initial speech retrieval interface; lexical retrieval, string-based entity correction, and the text generator require textual inputs. The deterministic speech embedding does not constitute a calibrated model of acoustic uncertainty. Reported retrieval results reach 87.6 Recall@10 for the speech-only retriever and 90.4 for the hybrid configuration, compared with 81.0 for transcript-first dense retrieval. Medical QA and ablations provide descriptive system comparisons, with input-provenance and evaluation limitations stated explicitly. We distinguish controlled synthetic acoustic tests from real clinical speech, qualify component attribution, and specify the remaining comparisons needed to assess speech-only QA and clinical generalization.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_f51106eff75111e5b9d75504