AlGrano: Voice Note Summary

The seven-minute voice note from the family group, the client who explains
everything out loud, the colleague who thinks while talking. AlGrano turns
it into three to five key points, your tasks, the dates mentioned and a
reply ready to send. On your iPhone, with no account and no microphone.

AlGrano is Spanish for getting straight to the point, and that is the whole
job:

1. SHARE THE AUDIO. In WhatsApp, press and hold the voice note, choose
Forward, tap the share icon and pick AlGrano. You can also import an audio
file straight from the app.

2. READ WHAT MATTERS. Three to five key points, the tasks aimed at you and
the dates that came up. The full transcript is right there too, in case you
want to check an exact sentence.

3. REPLY WITHOUT AGONISING. Three reply drafts with the same content in
different tones: diplomatic, neutral, firm. Tap the one that fits, it lands
on the clipboard, you paste it into the chat and send it yourself. The app
never sends anything.

To-dos move into Apple's Reminders with one tap, and "Thursday at six"
becomes a date you can hand to the iPhone calendar.

NOT A TRANSCRIPTION APP

WhatsApp already transcribes its own voice notes. A wall of unpunctuated
text still does not tell you what to do about it. That is where AlGrano
starts: structure and a reply.

YOUR VOICE NOTE NEVER LEAVES THE IPHONE

A personal voice note often holds more than you would tell a server: a
diagnosis, a salary offer, a falling-out with a business partner. AlGrano
uploads none of it. Transcription uses Apple's on-device speech recognition
and the summary uses the language model already shipped inside iOS. There is
no server receiving your audio, because there is no server. The app has no
record button and never asks for microphone access; audio only comes in
through the share sheet or a file you import. No account, no analytics, no
ads, no cloud sync. Once the speech files are in place (a one-time download),
everything works in airplane mode.

THE AI SUGGESTS, YOU CHECK AND SEND

The model summarising your audio is small and runs on a phone battery. It
gets most things right and some things wrong: it can mishear a name or put a
date in the wrong week. That is why every result is marked as an AI
suggestion and the full transcript is always one tap away, so you can read
what was actually said. The three replies are drafts; you send them from
your own chat, with the changes you see fit.

WHAT IT COSTS

Nothing. Sharing a voice note and getting the key points, to-dos, dates and
the three replies is free, with no subscription, no paywall and no monthly
quota. If the app saves you a few six-minute listens, you can leave a tip
from the settings. Tips are voluntary and unlock nothing.

REQUIREMENTS

AlGrano needs an iPhone that supports Apple Intelligence, running iOS 26,
with Apple Intelligence enabled in Settings. The language model is part of
iOS and needs no key; the speech files for German, Spanish and English are a
one-time download the app asks you for.

Questions or problems: support@freshlab.es. We answer in English, Spanish
and German.