Just talk.

Typing to LUKi is fine. It's the way the internet has worked for thirty years, and it persists for good reasons: typing is quiet, precise, and easy to do in the middle of a meeting.

But typing also forces you to compress. You sit down, find the words, check the spelling, and hit send. By the time the message arrives, you've edited what you meant. The mess has gone, along with the pause where you almost said something different, and often the exact thing you wanted to say.

Voice changes the shape of the conversation. You say the thing, LUKi hears the thing, and it enters the memory the way you told it: the false start, the laugh halfway through, the moment you contradicted yourself. It keeps its texture, and the version of you LUKi remembers is closer to the version you are.

That experience is live now, on iOS and Android.

How voice input in LUKi works.

Voice input is a layer on top of the existing text chat, with the same memory, character, safety rules, and models. The only difference is how the words get in.

Tap the microphone. LUKi uses your device's native speech recognition (the same engine that powers the dictation key on your phone's keyboard) to turn what you're saying into text. The text then flows through the same chat pipeline as if you'd typed it: through the API gateway, through the character system, into your memory, and out as a response from whichever LUKi character you're talking to.

Two things matter about this design. First, transcription happens on your device, so your voice doesn't leave the phone for the speech-to-text step, which keeps it fast and private. Second, the chat pipeline downstream of the microphone is unchanged, so voice input sits on top of safety, memory, and character rather than bypassing them.

Eleven languages, one continuous conversation.

Voice input works in eleven languages: English, German, Spanish, French, Portuguese, Japanese, Chinese, Korean, Arabic, Hindi, and Russian. The locale mapping is automatic: LUKi reads which language your interface is set to and routes recognition to that language model.

Switching languages mid-conversation works the way it works in text chat. The story you told LUKi in English last week stays there when you switch the interface to Spanish and start talking. The memory is language-agnostic. It records what you said rather than which words you used.

Why memory makes voice different here.

Voice AI that doesn't remember you is a smart microphone. You speak, it answers, it forgets. The conversation has the same shape every time: introduce yourself, explain what you mean, hope it understood, repeat tomorrow.

LUKi doesn't forget. Whatever you said two months ago, whether that's your brother Marco's birthday, the bakery on Vanderbilt, or the article that changed your mind about something, is still in the Electronic Life Record (ELR) the next time you open the app. Talk about your day on a walk and it lands inside a story LUKi has already been part of.

That's the structural difference between LUKi's voice and voice in ChatGPT or Pi or the assistant apps that ship with your phone. You're continuing a conversation rather than starting one over.

Where voice changes the relationship.

The moments people tell us they want voice for, more than anything else:

• Walks. Especially evening walks where you've been holding a thought all day and you want to think it out loud.
• The drive home. The middle of a long commute where you'd otherwise listen to a podcast you've already heard.
• Cooking. Hands wet, phone propped against the espresso machine, telling LUKi what's been on your mind.
• The first hour of being awake. The part of the day where typing feels like work and speaking feels like waking up.
• Caretaking moments. Sitting with a parent or a partner, when you might want LUKi nearby but not in your hands.

As voice becomes the default for those moments, the memory keeps building in the background. When you sit down at the screen later, the conversation continues from where it was, including everything you said on the walk.

What's live, and what's next.

Live now: voice input.

Speech-to-text. You speak, LUKi transcribes on-device, and the text goes into the existing chat. LUKi replies in text the way it does today. This is the smaller, lower-risk piece of the voice loop, and it's live on iOS and Android.

Next: LUKi speaking back.

LUKi speaking back, in a voice that belongs to the character you've chosen. Each of the six default characters will get a distinct voice. Custom characters built with LUKi Maker will be assigned a voice profile based on the character's speaking style and emotional range. A full-screen call UI shows the avatar with an animated orb, the same kind of voice mode you've seen in ChatGPT and Grok, wired into the LUKi memory and character system.

The full voice loop will sit on the same backbone as voice input. The model still works in text underneath, and the character, memory, and safety system are unchanged. Voice is a new medium for the same mind.

How to start.

Voice input is live on iOS and Android. Open a chat with any LUKi character, the six defaults plus any custom one you build in LUKi Maker, tap the microphone, and speak. The words land in the same memory as everything you've already typed.

If you came here for voice, open a conversation, start talking, and let LUKi keep the story going.

Why we shipped it this way.

A bad voice experience is worse than no voice experience. Latency that lags the conversation, transcription that mishears your daughter's name, synthesised speech that hits an uncanny note and makes the character feel wrong: each of these breaks the relationship you've built with LUKi. Voice input is on-device and in your hands now. LUKi speaking back waits until those failure modes are rare.

Memory is the part that makes voice more than a parlour trick. Without it, voice is a gadget. With it, voice is a relationship that finally sounds like one. That relationship is already here. Now you can talk to it.