One short line per turn
Every time someone finishes speaking, Akoe rewrites what they said as a single line. Not a transcript that scrolls away while you are still reading the first half.
Live captions for iPhone
Akoe listens to the conversation around you and shows what each person said as one short line, in Spanish or English, whichever they used. Big enough to read. Short enough to keep up.
Will calls her tomorrow before 10.
Yeah, I'll ring her tomorrow morning, before ten, so she knows we're coming.
Dile que llevamos el postre.
Vale, pero dile también que llevamos el postre, que el año pasado no llevamos nada y quedó feo.
Dad, are you coming with us?
And Dad, do you want to come with us on Sunday or would you rather
Why another captioning app
A running transcript of a family dinner is a wall of text. By the time you have read a sentence, three more have arrived.
Most captions do not say who is talking. In a room with four people, that is half the meaning gone.
Families that mix Spanish and English do not stop to switch a language toggle. The app should not need them to.
Features
Every detail is there so you can read the line before the next one arrives.
Every time someone finishes speaking, Akoe rewrites what they said as a single line. Not a transcript that scrolls away while you are still reading the first half.
The short line is big. Underneath, in smaller type, are the words as they were actually said. Glance at one, check the other. Nothing is hidden.
Recognition switches language automatically, mid-sentence. Nobody has to pick a language first, and nobody has to remember to switch it back.
Each person at the table gets a color and a label. Tap it once to give them a name, and the name sticks for the rest of the conversation.
One button. The last two minutes of conversation come back as three lines, so you can rejoin without asking anyone to repeat themselves.
Akoe learns your voice the first time you open it and keeps your turns off the feed. What you read is what other people said.
A text size control on top of the phone-wide Dynamic Type setting, up to one and a half times. Bold Text, Increase Contrast and VoiceOver all respected.
New lines push older ones up. The Catch me up button never moves. You never have to chase a button around the screen.
Speech recognition runs on the iPhone itself. Only text goes to the model that writes the short lines, and on iPhone 15 Pro or newer even that stays on the device.
How it works
Apple's on-device speech recognition turns voices into words, in Spanish and English, switching on its own. This runs entirely on the phone and costs nothing.
Akoe splits the stream into turns on the pauses between speakers, gives each voice a color, and drops your own voice so the feed is only what others said.
The words of each turn are rewritten as one short line, about a second after the person stops. The original words appear instantly underneath, so you are never waiting on a blank screen.
Never audio. On iPhone 15 Pro and newer, nothing at all: Apple's on-device model writes the short lines too. On older iPhones, only the text of each turn is sent to a small AI model and comes straight back. Nothing is stored anywhere.
Catch me up
The button at the bottom of the screen never moves. Press it and the last two minutes of conversation come back as three lines: who is doing what, and what they are asking you. Then straight back to live.
Sunday: visiting your sister. Mamá wants her called first.
Will calls her tomorrow morning. You're bringing dessert.
They're asking if you want to come.
Compare
The honest version: if Apple's built-in Live Captions work for you, use them, they are free. Akoe is for when they don't.
| Feature | Akoe | Apple Live Captions | Ava | Otter |
|---|---|---|---|---|
| Made for a conversation in the room | Yes | Yes | Yes | Meetings and recordings |
| Short lines instead of a running transcript | One line per turn | Running transcript | Running transcript | Running transcript |
| Spanish and English switching mid-sentence | Automatic | One language at a time | One language per session | Mainly English |
| Tells voices apart | Color and name | No | Yes, by color | Yes |
| Two-minute catch-up on demand | Yes | No | No | Notes after the meeting |
| Keeps your own voice off the screen | Yes | No | No | No |
| Audio stays on the phone | Yes, text only leaves | Yes | Cloud | Cloud |
| No account, no sign-up | Yes | Yes | Account required | Account required |
| Price | €4.99 / month | Free, built in | Free with 40-minute sessions; $14.99 / month | Free for 300 minutes a month; $16.99 / month |
Competitor details from their public pages, September 2026. Prices in the currency each one lists. Google's Live Transcribe is not included because it is Android only.
Pricing
€4.99 / month
Billed through the App Store. No account, no card details in the app.
Coming soon to the App Store
Launch pricing. Free for a week, so you can try it at a real meal first.
Questions
Apple’s Live Captions are free and good, and if they work for you, use them. They show a running transcript in one language, with no idea who is speaking. Akoe is built for the family table: one short line per person, colors and names for each voice, Spanish and English mixed freely, and a Catch me up button for when you lose the thread.
No. Nothing is saved. Akoe keeps the last two minutes in memory so that Catch me up can work, and forgets them as the conversation moves on. When you close the app, it is gone.
Nowhere. Speech recognition happens on the iPhone using Apple’s on-device engine. The text of each turn is sent to an AI model that rewrites it as one short line, and comes straight back. On iPhone 15 Pro and newer, Apple’s on-device model does this step too, so nothing leaves the phone at all.
Any iPhone that runs iOS 26, which means iPhone 11 or newer. Fully on-device short lines need iPhone 15 Pro or newer. iPhone only; there is no iPad or Android version.
Spanish and English, including sentences that start in one and finish in the other. You can choose which language the short lines are written in, whatever language was spoken.
Akoe splits turns on pauses, so overlapping voices are the hardest case for it, as they are for every captioning app. The original words are always shown underneath the short line so you can see exactly what was picked up.
No. Akoe is for people in the same room: meals, visitors, the doctor, the bank. For calls, your iPhone already has built-in call captioning.
No. Download it, read two sentences so it learns your voice, and put the phone on the table. There is nothing to sign up for.
Akoe was built for one person's dad. It is now for anyone who has ever nodded along and hoped the question wasn't for them.