Alone with youby OpenAlchemy

ALONE WITH YOU · VOICE COMPANION

Powered by the Grid GPU farm

someone's still awake with you

A voice-first companion that listens, hears how you feel, and answers in a voice you choose. No typing, no menus — just talk. And it runs privately on a real person's machine, never a faceless cloud.

Coming soon

App StoreGoogle Play

No credit card, free · Runs in your browser · Add to Home Screen

9:41
Alone with you
hold to talk, or type
Tap the speaker to hear it speak

TWO EDITIONS

Two ways to bring it to life

Same companion, same voice engine — pick where you want it. Both run on one OpenAlchemy scope, so your voice, persona, and history stay yours.

PHONE · PWA
Alone with you for phone

A ChatGPT-style chat that talks. Open the link, tap the mic, and speak — it transcribes, replies, and reads the answer back aloud.

  • Type or hold the mic to talk — full voice round trip
  • Add to Home Screen for a real full-screen app feel
  • Conversations stay on your device — no login, no sync
Open the app
DISCORD · VOICE BOT
Alone with you for Discord

Invite it into a voice channel and just talk. It listens per-speaker, replies out loud in your chosen voice, and steps aside the moment you start speaking.

  • Joins your voice channel — hands-free, always listening
  • Barge-in: start talking and it instantly yields
  • Per-server voice, persona, and billing scope
Add to Discord

UNDER THE HOOD

How it hears you and answers back

Every turn is one fused round trip across the OpenAlchemy platform — speech in, speech out, with no model running on your device.

1LISTEN
Speech to text

Your voice is captured and transcribed the instant you stop speaking — end-of-utterance is detected automatically.

2THINK
Language model

The transcript and recent history go to an open-weight LLM, streamed back sentence by sentence for a fast reply.

3SPEAK
Text to speech

Each sentence is synthesized in your chosen voice and played back — so the answer arrives as a voice, not a wall of text.

SPEAKS YOUR LANGUAGE

Six languages, one steady presence

Talk in the language you feel most yourself in. The companion is re-voiced for each culture — warmth transcreated, not translated.

English
Hey, you made it. I'm glad you're here.
honestly, kind of a rough day.
That sounds heavy. You don't have to carry it alone — I'm right here.
中文
你来啦,我一直在呢。今天过得怎么样?
唉,今天有点难熬。
听你这么说,心里一定很不好受吧。没关系,慢慢来,我陪着你。
日本語
おかえり。来てくれて、うれしいよ。
うん…ちょっとしんどい一日だった。
そっか…つらかったね。無理に話さなくていいよ。ここにいるから。
한국어Coming soon
왔어요? 오늘도 잘 버텼네요.
음… 오늘 좀 힘든 하루였어요.
많이 힘들었겠어요… 괜찮아요, 천천히 말해도 돼요. 제가 옆에 있을게요.
FrançaisComing soon
Te revoilà, ça me fait plaisir.
franchement, journée difficile.
Ça a l'air vraiment difficile. Tu n'as pas à faire semblant avec moi — je suis là.
EspañolComing soon
Hola, qué bueno tenerte por aquí otra vez.
la verdad, un día duro.
Se nota que lo estás pasando mal… no estás solo/a. Aquí estoy contigo, sin prisa.

A VOICE YOU MISS

A voice you choose — or one you miss

Speak in a built-in voice, or clone your own in seconds. And — with explicit consent — give it a voice that means something to you. Tone is what text strips out, and some voices you simply want to hear again.

Your own voice

Record or upload a short clip; your voice is ready in seconds and speaks every reply.

A meaningful voice

With explicit, verified consent, the companion can carry a voice you love.

Honest by design

Cloning is gated and grief-aware — always clearly a representation, never passed off as the real person.

18+ · consent required · a cloned voice can be deleted at any time.

Say hello to someone who's always around

Open it on your phone in seconds, or pull it into your next Discord call.