Memo iOS
Support
Memo turns a spoken thought into a written journal entry. Speak, and it transcribes on your iPhone and saves plain Markdown files you own.
Email hi@santiagoalonso.com. I aim to reply within a few working days.
Getting started
Open Memo and complete the short setup: choose what kind of entry you want, which transcription model to use, and which AI writes the entry. Memo downloads its transcription model at the end of setup, which happens once and needs a connection. Then tap the large record button, speak freely, and tap finish.
Where are my entries saved?
By default in Files, under On My iPhone, Memo. You can change this in Settings, Folder to any Files or iCloud Drive folder, or to Google Drive. Changing it only affects future entries; existing files are never moved.
Does Memo work offline?
Yes, once the transcription model is downloaded and you are using the default on-device processing. Recording and transcription both run entirely on your iPhone. The internet is only needed to download a model or to use an optional cloud provider.
Does my voice recording leave my iPhone?
No. Audio is never uploaded in any configuration. If you choose a cloud provider, only the text transcript and your instructions are sent.
Can Memo handle languages other than English?
Yes, if you select the Most Accurate transcription model, marked Multilingual in Settings, Transcription. It detects the language you speak on its own. The Fast model is English only.
Do I need to pay for an AI provider?
No. Memo uses Apple's on-device intelligence by default, which is free and private. Cloud providers are optional and use your own API key, billed by that provider. A consumer ChatGPT or Claude subscription does not include API access.
Can I choose which model a provider uses?
Yes. In Settings, Processing, pick a provider and then any model it offers. The list is fetched from the provider, and you can type a model ID directly if you want one that is brand new.
Can I redo an entry with a different model?
Yes. Open a completed entry and use Reprocess at the bottom. It reuses the saved transcript, so your audio is not needed and is not re-sent. Existing exported files are left untouched.
The first recording is taking a long time
Preparing the Most Accurate model for the first time can take a few minutes because it is large. Later recordings start immediately. The Fast model prepares in seconds.
Something went wrong during processing
Your recording is kept. Open the entry and choose retry. Memo continues from the last completed step rather than starting over.
How do I run setup again?
Settings, About, Run setup again. Nothing is reset until you change it.
Reporting a problem
Email what you expected and what happened instead, your iPhone model and iOS version, the Memo version from Settings, About, and which transcription and processing models were selected. Please do not include the contents of a private entry unless it is needed to explain the problem.
Privacy
Read the Memo privacy policy.