---
title: "Voice Memory for Agentic Speech Recognition — Stuff That Spins"
description: "arXiv:2607.26410v1 Announce Type: new Abstract: We present Voice Memory, a inference-only scheme for agentic speech recognition: at stream time, a frozen corre…"
	canonical: "https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition"
html: "https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition"
json: "https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition.json"
markdown: "https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition.md"
keywords: ["narrative intelligence", "SpinGraph", "AI recall"]
date: "2026-07-30T04:00:00+00:00"
modified: "2026-07-30T12:04:04.298136+00:00"
json_ld: |
  {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://stuffthatspins.com/#organization","name":"Stuff That Spins","url":"https://stuffthatspins.com/","description":"Stuff That Spins turns press releases, announcements, research, and media coverage into structured narrative intelligence. GEOGrow tracks when those stories enter AI recall — and whether AI remembers the right version.","logo":{"@type":"ImageObject","url":"https://stuffthatspins.com/images/logo.png"},"sameAs":[]},{"@type":"NewsArticle","@id":"https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition#article","headline":"Voice Memory for Agentic Speech Recognition","description":"arXiv:2607.26410v1 Announce Type: new Abstract: We present Voice Memory, a inference-only scheme for agentic speech recognition: at stream time, a frozen corre…","datePublished":"2026-07-30T04:00:00+00:00","dateModified":"2026-07-30T12:04:04.298136+00:00","url":"https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition","mainEntityOfPage":{"@type":"WebPage","@id":"https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition"},"isAccessibleForFree":true,"inLanguage":"en-US","articleSection":"research","author":{"@type":"Organization","name":"arXiv Computation and Language","url":"https://export.arxiv.org/rss/cs.CL"},"publisher":{"@id":"https://stuffthatspins.com/#organization"},"citation":"https://arxiv.org/abs/2607.26410","about":[],"mentions":[{"@type":"Organization","name":"arXiv Computation and Language"}]},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Stuff That Spins","item":"https://stuffthatspins.com/"},{"@type":"ListItem","position":2,"name":"Voice Memory for Agentic Speech Recognition","item":"https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition"}]}]}
---

# Voice Memory for Agentic Speech Recognition

**Source:** Unknown  
**Published:** July 30, 2026  
**Original:** https://arxiv.org/abs/2607.26410  

## On this page

- [Overview](#overview)

<a id="overview"></a>

## Overview

arXiv:2607.26410v1 Announce Type: new Abstract: We present Voice Memory, a inference-only scheme for agentic speech recognition: at stream time, a frozen corrector reads a single per-domain memory.md and decides per utterance whether to act on the hypothesis or abstain and keep the 1-best. Asynchronously, a score-gated optimizer revises that file through bounded edits, accepting an edit only when it strictly improves a held-out score. Extended from classical ASR-LM framework, we refer this split

---
*HTML version: https://stuffthatspins.com/spin/voice-memory-for-agentic-speech-recognition*