aloud.

your voice is an overpowered and underrated tool for meditation and inner work.

a meditation facilitator that listens and responds. speak what you're noticing, and it asks thoughtful questions to help you explore.

here, feeling good is aloud.

open the appbeta download how it works

browser · macOS · Windows · Linux · free & open source

subscribe to my substack or follow me on Twitter for news, including about iOS and Android.

two ways to practice

exploration

a dyadic format. you speak freely about what you're experiencing in the moment. aloud. asks questions to help you go deeper, play, or whatever your intention is.

mix and match attention focuses - body sensations, emotions, or parts work / IFS - with vibes like playful, compassionate, or spacious. you can customize the guidance to be more or less chatty or directive.

Why? It's fun! It lets you explore your own mind and body in an open-ended way that helps you make connections. It can also be useful for learning how to enter the jhanas.

noting

name something in your awareness in one or two words, like "thinking" or "warmth". then other participants that you choose - AIs, fixed phrases, or sound effects - each take a turn in order, until it's your turn again.

Why? This is a very pure way to experience rest → cue → feel inward → notice → speak which is a surprisingly pleasant and potent stabilizer of attention. Rest lets you settle, while speaking helps keep you alert.

Of course, vibes are best with human partners, so that's encouraged if possible.

aloud's main screen — a central orb with controls around it

bring your own model

aloud. works with whatever LLM you prefer and makes setup easy.

fully localollama, nothing leaves your machine. can install for you.
cloud APIsanthropic, openai, openrouter, venice, groq (has free tier)
claude subscriptionuses claude CLI/Claude Code (mac/win/linux)

or let the aloud. cloud handle it.

just open the app, sign in, and start. we relay to hand-picked AI and voice providers, so there's nothing to install and no keys to wrangle. new accounts get free credits, then you top up prepaid credits only as you go.

open the appbeta optional & in early access · privacy · terms

aloud is open source.

no paywall, no tracking. run it yourself and no account is needed. the source is on GitHub and you can build it yourself in a minute.

if it helps you and you're able, chip in to help keep it going. your tip pays for the Apple developer license, hosting, and updates.

ko-fi venmo kind words

download

latest release: checking…

macOS
.dmg installer
drag to Applications
Windows
.exe installer
double-click to install
Linux
.AppImage
chmod +x, double-click or run

don't want to install anything? open aloud right in your browser (early access). prefer to build from source? it's a clone + one script.

questions

why make this? you don't need an app to meditate.

I've had amazing experiences with practices like the ones in this app and wanted to make them accessible to everyone. exploration mode resembles both human-guided somatic exploration and the AI-powered meditation I did on my Jhourney retreat; I've had light jhana experiences with both. I've also done a lot of group noting sessions and find that it helps me go deeper than I would usually on my own. it's best with humans but most of the mechanics still work well here.

there are also many worthwhile inner-work and meditative practices where an app would just be a distraction; please do explore further.

why out loud?

speaking is really helpful in maintaining alertness, and enforces the practice of choosing a word (noting mode) or longer bit of speech (exploration mode) for your experience; this helps build your internal vocabulary, and a better internal vocabulary lets you understand yourself better.

plus, in my experience there's something special about voice-based practice that I don't have a complete explanation for; I just seem to have deeper experiences when my mind and body are being engaged in that way.

how do I start - do I need to install, sign up, or pay?

running aloud. on your own device needs no account and costs nothing unless you choose to use aloud cloud for convenience. aloud cloud uses an account and payment system to provide text-to-speech, speech-to-text, and AI facilitator brains; you'll receive a few free credits when connecting a Google or Apple account. you can always use your own device's capabilities or API keys instead of paying.

the web app needs no installation and runs right in your browser at aloud.rest/app. for technical reasons, the web app cannot connect to services in the same way as the desktop app does. so, depending on your browser and device, aloud cloud may be your only option for text-to-speech, speech-to-text, and AI brains on the web app.

aloud. is open source and free to run; tips are deeply appreciated - hundreds of hours of care have gone into making this app the best it can be. tips also help offset expenses like developer licenses, hosting, and signup credits.

why AI? won't it feel cold or too sycophantic?

in noting mode, AIs are optional and just say one or two words at a time.

exploration mode by nature needs a partner that can give useful responses. the prompts have been carefully crafted to reflect the idea that the AI is here to be a facilitator for your own exploration, not to tell you what to do or believe. AI is not perfect, but in my experience this works pretty well. the newer models even have enough emotional intelligence to sometimes suggest useful avenues of inquiry.

you always have full freedom in choice of LLM.

what's parts work / IFS?

Internal Family Systems (IFS) is a popular framing of parts work - in parts work, the mind is seen as having a collection of "parts" rather than a single voice. you've probably said something like "part of me wants to be productive but another part just wants to binge Netflix" - this just takes that idea seriously. it can be a really powerful way to investigate internal conflict or deeper emotions.

aloud. can be a partner for your parts work if you choose the "parts" focus.

are my conversations or other data stored anywhere?

your session transcripts are saved on your own device (in the desktop app, or in your browser for the web app), and you can opt not to save them, or delete them later. if you use a cloud AI model, your words go to that provider to generate a response, under their terms. if you choose cloud text-to-speech, it only receives the AI's responses, not your words.

when you run aloud. and use your own AI/TTS/STT, you don't need an account and there's no server in the middle. aloud cloud, the optional hosted service, relays your session to service providers but doesn't store any of it. see privacy for details.

is this therapy?

no. it's a practice tool for meditation and self-exploration, not a substitute for an experienced human therapist. if something difficult or overwhelming comes up, please don't push it and please do reach out to a professional for support.

with gratitude to Maija Haavisto, Jhourney (who thank Doug Kraft), and Vince Horn for guiding me in similar practices.