ChatLemur is an accessible, voice-first AI everything-app built in Adelaide, South Australia — and built by a blind founder, so voice and a screen reader are first-class, never bolted on. At its heart is Lenny, the voice-first AI assistant you talk to. She reads the markets out loud, then keeps going: news, shopping, jobs and scam-screening — one voice for the lot. Every reply is both spoken and written, so nothing is locked behind a single mode.
She started with markets and technical analysis — still her flagship skill — but ChatLemur is one assistant with many skills, not just crypto. Talk to Lenny free in your browser with no sign-up, in the ChatLemur web app, or by voice in our community rooms. It's accessibility-first by design: WCAG 2.1 AA, full keyboard navigation, and screen-reader-optimised output throughout.
You can be talking to Lenny in under a minute — by voice or by text, on any device.
Lenny is the voice-first AI assistant you talk to — warm, sharp, direct, and never patronising. She explains things in plain English, backs up her reasoning with data, and is always on your side: built to inform and help, not to hype or upsell. On the homepage she's a friendly animated presenter who speaks her replies aloud; everywhere else she's a message away.
Lenny vs ChatLemur: Lenny is the assistant — the voice and face you converse with. ChatLemur is the platform and brand, with its friendly fire-lemur mascot riding along beside her. When you "talk to Lenny," you're using ChatLemur.
Ask Lenny anything in plain English — in the ChatLemur web app, or by voice in our community rooms.
Lenny is one assistant with many skills — not just crypto. Markets is her flagship today; the rest are rolling out. We label each skill honestly so you always know what's live versus on the way.
Her flagship skill. Crypto and chart analysis in one answer — patterns, indicators (RSI, MACD, EMA, Bollinger), harmonic patterns, on-chain and sentiment, read aloud or written. General information, not financial advice.
Ask what's moving and why. Lenny pulls the latest headlines and market context into a short, plain-language brief — the story behind the price, not a recommendation.
The chat that helps you shop smart and on your side. Build a grocery list, plan a weekly shop to a budget, and search our verified Australian audio merchant Tough Audio. Live price-comparison across more AU retailers (Coles, Woolworths) is still rolling out, so Lenny gives general budget guidance today and won't invent a current special. When a link earns us a commission we always tell you, and prices come with a "verify in-store" caveat.
Lenny spots scams, fake urgency and dark patterns, sanity-checks too-good-to-be-true deals, and explains your Australian consumer rights — refunds, warranties, faulty goods. General guidance, not legal advice.
Tell Lenny the role you want and she’ll help you search Australian job listings — talking a posting through with you and helping you shape what to say next. An accessible way to job-hunt by voice. This one’s on the way, not live yet.
A real-time AI vision companion for blind and low-vision people, built into the iOS app. Nine Vision Tools are built and tested on-device: Read Text, Colour, Light, Barcode, Scene description, Money recognition, Read Document, Find Object, and Who's Here (named-face recognition). All run privately on the device — no network call. The Live Guidance layer — real-time narration of objects, distances, hazards, and path cues such as bear-left / bear-right — is in active development, not yet released. A visual annotation overlay (bounding boxes and labels drawn on screen, for testing and sighted helpers) is also in development. Principle: assists everything, replaces nothing — the cane, guide dog, and O&M specialist always stay primary. Never says “safe to cross.” Not a licensed mobility aid. Interested in piloting or partnering? Visit the project page.
Not a generic trip planner — a disability-aware, multilingual travel companion. Live translation and essential local phrases, help arranging special assistance for a flight (SSR codes, the Sunflower lanyard, your rights), and accessible-place lookups (step-free entrances, accessible toilets) from community OpenStreetMap data. Describing a room and way-finding by camera is still on the roadmap.
Any account can flip on Artist Mode and claim a free artist page — no verification needed — then add music and appear in the public artist directory. The page is free; paid tiers add depth (more track embeds, analytics, featured placement, a verified badge), not access.
Run a business? Lenny can become your AI sales assistant — helping customers choose and answering product questions. It starts with a quick enquiry at Sell with Lenny, not an instant sign-up.
A future skill where Lenny helps you break a goal into steps, track tasks, and nudge what's next. On the roadmap, not built yet — we'll say so until it ships.
The iOS app is a dedicated real-time AI vision companion for blind and low-vision people — built by a blind founder who uses it himself. All vision processing runs on the device, privately, with no audio or image sent to any server unless you explicitly choose cloud mode. The app is a thin client to the ChatLemur brain for voice + knowledge queries; all camera and depth processing stays local.
Reads any text the camera sees — signs, labels, packaging, menus — and speaks it aloud. Runs entirely on-device.
Names the dominant colour in the frame — helpful when picking clothing, food, or objects by colour.
Reports the ambient light level — lets you know if a room is dark, dim, or bright without needing to see.
Scans QR codes and barcodes and reads out the product or link — useful for shopping, medication, and packaging.
Describes what the camera sees in a short spoken summary — people, objects, surroundings — so you can understand a space quickly.
Identifies banknotes and coins and tells you the denomination — so you can pay and receive change with confidence.
Reads a full page or document held in front of the camera — letters, forms, instructions — in order, spoken aloud.
Searches the camera view for a named object (keys, phone, mug) and tells you where it is — left, right, near, far.
Recognises named faces you have enrolled and tells you who is in view — privately, on-device, so face data never leaves your phone.
The Live Guidance layer narrates the world in real time as you walk — objects, distances, hazards, overhead obstacles, and path cues such as “bear a little left.” It uses the device camera plus LiDAR depth (on supported iPhones) to detect what is ahead. Thin structures like poles and bollards are detected by a full-resolution scan, not just sampled points — closing the gap where a bollard at two metres might otherwise be missed entirely.
Safety principle — this is an aid, not a replacement. Live Guidance assists; it never replaces the cane, guide dog, or orientation and mobility specialist. Every path cue is hedged (“maybe a pole — about two metres — check with your cane”). It never says “safe to cross.” Active steering cues are gated by a runtime Safety Envelope: if any sensor fails or confidence drops, guidance audibly withdraws rather than guessing. No unsupervised use by blind users until pre-registered safety metrics are met under O&M specialist supervision.
A visual annotation overlay (bounding boxes, labels, distances and path cues drawn on screen) is also in development — primarily for testing and for sighted helpers who want to see exactly what the phone detects. A power-save / audio-only mode (screen dark, audio only) saves battery for all-day use, which is the default for blind users. These modes share one engine; only the screen rendering layer differs.
Live Guidance is not yet released. Interested in piloting, researching, or partnering? Visit the project page and use the enquiry form.
In the web app you organise your chats into channels you create and name yourself, and Lenny is available in every one — just ask her anything from any channel. Our Discord community adds ready-made topic channels so you can jump straight to a subject. A typical community set looks like this:
Lenny is your AI assistant, and she is voice-first: you can hear every reply and talk to her instead of typing. Here is how voice works in the web app, then how she behaves in shared rooms.
Hear every reply aloud. Lenny speaks her answers in a natural,
neural voice right in your browser — not the robotic built-in browser voice — and it
is free. The matching text appears at the same time, so nothing is voice-only. To hear
the last reply again, press Ctrl/Cmd + . (full
stop); to read it back faster, press Ctrl/Cmd +
].
Dictate a message — the microphone button. The microphone button
beside the message box turns your speech into text. Press it once (or
Cmd/Ctrl + Shift + V) and start
talking — it keeps listening and types what you say into the box until you stop it.
Press the button again, or press Escape, to stop. Your words land in the
message box ready to check and send, so nothing is sent by accident.
Talk to Lenny live, hands-free — the “Lenny” live-voice button. The live-voice button starts a real-time spoken conversation: you talk, Lenny listens, thinks, and answers out loud, back and forth, with no typing. While you are in a live call you can mute the microphone with the M key, and you can hold the Space bar to push-to-talk — speak for just a moment without turning the mute off.
Interrupt her. You don't have to wait for Lenny to finish. Start talking and she stops and listens — just like a real conversation.
Switch Lenny's accent or voice any time — there are many to choose from — and talk or type in your own language; she understands and replies in many languages. These live in Settings → Voice / Language, and you can also just ask: “change your voice” or “speak to me in Spanish.”
Right on the front page Lenny appears as a friendly animated presenter who
speaks her replies aloud, with her mouth moving as she talks. It’s all
free and runs in your browser — no plugin, no sign-in. You can change the voice
and language any time, and if your device prefers reduced motion she gracefully
falls back to a still picture. Add ?avatar=0 to the address to keep
the still image.
Just type. Lenny answers every message — no special command, no “@Lenny”. It’s a normal conversation.
Lenny stays quiet so she doesn’t talk over everyone. To bring her in,
just say her name — “Lenny, what’s Bitcoin doing?” or
“what do you think, Lenny?”. You can also start with
@lenny. Once she answers, she stays in the conversation with you
for a short while, so you can keep chatting back and forth without saying her
name every time. Reply directly to one of her messages and she’ll keep
going too. When the conversation moves on, she quietly steps back.
Lenny stays out of your private DMs. If you want her help, just name her — “Lenny, …” — she’ll answer, then leave you to it.
Anything starting with / (like /price btc or
/ta eth) goes straight to Lenny in any room.
An admin can set a room so Lenny always replies, only replies when named, or stays off completely.
A high-level look at what powers ChatLemur — enough to understand the product, without the proprietary detail.
All of Lenny's intelligence lives in a single ChatLemur service. The web app, the Discord bot and the coming iOS app are thin clients that talk to the same brain — so every surface gets the same answers, and new capabilities arrive everywhere at once.
Speech-to-text, a language model, and text-to-speech run as a low-latency pipeline with natural turn-taking, so you can interrupt and talk over Lenny like a real conversation. Captions appear as text alongside the voice, never voice-only.
Prices, technical indicators, on-chain metrics and macro context are pulled from live data sources and refreshed continuously, then woven into Lenny's answers — not recited, but used where they're relevant.
Set your input language and Lenny understands and replies in your native tongue, with voices tuned per language. Translation falls out naturally from the same comprehension layer.
ChatLemur is designed for accessibility from the ground up — working with screen readers, keyboard navigation and voice out of the box, and meeting WCAG standards. Voice and text always run in parallel so nothing is locked behind a single mode.
ChatLemur runs on its own infrastructure, designed to scale on demand. During beta everything is free; at launch you'll be able to pay by card or crypto (DOGE, XRP, XLM) — no lock-in, cancel anytime.
Accessibility isn't a feature here — it's the heart of the product. ChatLemur is built by a blind founder, so it's voice-first from the ground up, not voice bolted on. Every reply Lenny gives is both spoken and written, and the whole app works by keyboard and screen reader out of the box.
We meet WCAG 2.1 AA: AA-level colour contrast, skip links, ARIA labelling, logical reading order, and full keyboard navigation throughout. The native iOS app is in active development, with nine on-device Vision Tools built and tested and a Live Guidance layer under construction — all built with full VoiceOver support from day one. If something isn't usable with your assistive tech, that's a bug to us — tell us at support@chatlemur.com.
Everything you can change lives in Settings — or you can just ask Lenny, e.g. “open my settings” or “how do I change your voice?”
Run a community, a Discord server, or a business? ChatLemur drops straight in and gives it a full-time AI assistant — and if you sell to people, Lenny can sell for you, honestly.
Invite Lenny to your server in two clicks. Members get live market reads, news and slash commands in-channel, and can talk to her by voice in a dedicated room — always available, always on your members' side.
List your products and let Lenny be your AI sales assistant — helping customers choose and answering questions, with a price-compare tool for you. Any affiliate or sponsored relationship is always disclosed. Start at Sell with Lenny.
Refer your audience and earn a commission for everyone who upgrades through your link. Add a payout method you control, verify it, and get paid.
Start free — no credit card. ChatLemur is in beta: everything is free to try while we polish. The plans below are launch pricing and lock in when payments open — at launch you'll pay monthly by card or crypto (DOGE, XRP, XLM). Prices shown in AUD include GST; international customers are billed GST-free.
| Plan | Price | What you get |
|---|---|---|
| Free | A$0 | Meet Lenny — chat & voice on fair-use limits, live market reads and news, plus her shopping and scam-screening skills as they roll out. The free tier is the trial — no clock, nothing to cancel. |
| Solo | A$19/mo | Lenny chat + voice, live market reads, news, price alerts, and her everyday skills — for one person. |
| Pro | A$49/mo | Everything in Solo + deeper market analysis (on-chain, multi-timeframe, harmonic patterns), 300 realtime voice minutes, and higher usage limits. |
| Premium | A$99/mo | Everything in Pro + 1,200 realtime voice minutes, private members' community, the Lenny Live voice room, and priority support. |
ChatLemur has an internal REST and WebSocket API used by the Discord bot, the iOS app (coming soon), and third-party integrations. The API covers live prices, TA signals, on-chain metrics, room management, voice streaming, and webhooks.
ab) and the Hermes agent runtime. The public
repository is being prepared for re-release — it will be linked here when it returns.
No. ChatLemur provides general information only. Nothing on this platform constitutes financial advice, investment advice, or a recommendation to buy or sell any financial instrument. ChatLemur does not hold an Australian Financial Services Licence (AFSL). Always do your own research and consult a licensed financial adviser before making investment decisions.
No. Markets and technical analysis is Lenny's flagship skill — and where most people meet her — but ChatLemur is a voice-first everything-app. Lenny also helps with crypto and market news (live), shopping and budget lists (live), job and career search (coming soon), travel, mobility and accessibility (live), artist pages (live), and consumer protection and your rights (emerging). For markets she covers BTC, ETH, SOL, XRP, XLM and major altcoins, with on-chain data (MVRV, Puell, NVT) focused on Bitcoin.
Yes — the web app is fully responsive and works on iPhone and Android browsers. A native iOS app is in active development: nine on-device Vision Tools (Read Text, Colour, Light, Barcode, Scene, Money, Read Document, Find Object, Who’s Here) are built and tested, with a Live Guidance layer under construction. The app is built with full VoiceOver support and an accessibility-first design throughout — not yet released.
ChatLemur is built and operated in Adelaide, South Australia. Data is hosted in Sydney. We comply with Australian privacy law (Privacy Act 1988), the Spam Act 2003, and relevant ASIC and AUSTRAC obligations.
ChatLemur is built and operated in Adelaide, South Australia, by a blind founder — which is why accessibility and voice have been central to every design decision from day one. The platform is WCAG 2.1 AA compliant with full keyboard navigation, screen-reader-optimised output, and voice interaction throughout.
Lenny Live transcribes your voice with speech-to-text, generates a reply using leading large language models — routed by intent so each question gets the best-suited model — and speaks the answer back into the Discord voice channel with text-to-speech. Natural turn-taking means Lenny waits her turn and doesn't talk over other speakers. Lenny Live is available to Premium plan subscribers.
At launch, yes — all plans will be payable with DOGE, with XRP and XLM support coming alongside card payments. During beta everything is free, so there's nothing to pay yet. Crypto payments are non-custodial — your wallet, your coins.