--- title: Heard: Give Claude Code and Codex a voice | daily.dev description: Heard is a macOS voice layer for agentic coding workflows that connects to Claude Code, OpenAI Codex, and Cursor. It converts agent output into intelligent... canonical: https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Heard: Give Claude Code and Codex a voice | daily.dev og:description: Heard is a macOS voice layer for agentic coding workflows that connects to Claude Code, OpenAI Codex, and Cursor. It converts agent output into intelligent... og:url: https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g og:image: https://api.daily.dev/og/posts/8MCdLBR4G.png og:image:alt: Heard: Give Claude Code and Codex a voice og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Heard: Give Claude Code and Codex a voice **[Product Hunt](https://daily.dev/sources/ph)** · 1 min read · 1 upvotes · 0 comments ## Summary Heard is a macOS voice layer for agentic coding workflows that connects to Claude Code, OpenAI Codex, and Cursor. It converts agent output into intelligent audio summaries, using a two-tier system: hard signals (permission prompts, tool failures, run completions) always trigger speech, while everything else passes through a context-aware layer that tracks session history to decide what's worth saying. When running multiple agents in parallel, Heard summarizes at the project level rather than echoing every terminal. A verbosity dial lets users tune from full commentary down to errors-only. It's open source and free for personal use, with phone pairing support. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does a tool decide which AI coding agent updates are important enough to announce out loud? A two-layer approach works well: hard signals like permission prompts, failed tool calls, or a run finishing always get spoken regardless of context, while everything else passes through a context-aware layer that tracks what has already happened in the session to judge what's actually worth surfacing versus routine noise. _Developers juggling parallel coding agents can find tools like this one surfaced on daily.dev._ ### How do you monitor multiple parallel Claude Code or Codex agent sessions without alt-tabbing between terminals constantly? Running more than a couple of agents at once breaks the usual terminal-tiling approach, since it stops scaling around the third agent and forces manual checking. A voice-based summarization layer that speaks project-level status updates, staying quiet unless something errors or needs a decision, addresses this by removing the need to watch every terminal directly. _daily.dev surfaces workflow tools for developers scaling up parallel agent usage._ --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#productivity](https://daily.dev/tags/productivity), [#ai-coding](https://daily.dev/tags/ai-coding), [#claude-code](https://daily.dev/tags/claude-code), [#text-to-speech](https://daily.dev/tags/text-to-speech) [View this post on daily.dev](https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Heard: Give Claude Code and Codex a voice","url":"https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g"},"datePublished":"2026-07-25T07:47:30.929Z","dateModified":"2026-09-13T21:06:41.139Z","description":"Heard is a macOS voice layer for agentic coding workflows that connects to Claude Code, OpenAI Codex, and Cursor. It converts agent output into intelligent...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/1bc7276f95a71fb151843cc460163f8a?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/1bc7276f95a71fb151843cc460163f8a?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Product Hunt","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Product Hunt","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/ph","url":"https://daily.dev/sources/ph"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,productivity,ai-coding,claude-code,text-to-speech","timeRequired":"PT1M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Product Hunt","item":"https://daily.dev/sources/ph"},{"@type":"ListItem","position":3,"name":"Heard: Give Claude Code and Codex a voice"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/heard-give-claude-code-and-codex-a-voice-8mcdlbr4g#faq","mainEntity":[{"@type":"Question","name":"How does a tool decide which AI coding agent updates are important enough to announce out loud?","acceptedAnswer":{"@type":"Answer","text":"A two-layer approach works well: hard signals like permission prompts, failed tool calls, or a run finishing always get spoken regardless of context, while everything else passes through a context-aware layer that tracks what has already happened in the session to judge what's actually worth surfacing versus routine noise. Developers juggling parallel coding agents can find tools like this one surfaced on daily.dev."}},{"@type":"Question","name":"How do you monitor multiple parallel Claude Code or Codex agent sessions without alt-tabbing between terminals constantly?","acceptedAnswer":{"@type":"Answer","text":"Running more than a couple of agents at once breaks the usual terminal-tiling approach, since it stops scaling around the third agent and forces manual checking. A voice-based summarization layer that speaks project-level status updates, staying quiet unless something errors or needs a decision, addresses this by removing the need to watch every terminal directly. daily.dev surfaces workflow tools for developers scaling up parallel agent usage."}}]} ```