How to Log Expenses by Voice on iPhone (One Sentence, Logged)
Quick answer
To log an expense by voice on iPhone, open a voice expense tracker, tap the mic, and say three things: the amount, what it was, and where. “Twelve dollars, lunch, at the deli” is enough. An AI model turns the sentence into a transaction with an amount, a merchant or note, a tag matched to your own list, and a date. You can cover a whole day in one recording, in any major language. Review the result, fix anything misheard with Edit, and fall back to the keypad when you have no connection.
The hardest part of tracking expenses is not the math. It is the thirty seconds between paying for something and deciding whether logging it is worth the effort. A form loses that contest most days. A sentence does not. Voice logging exists so that "twelve dollars, lunch, at the deli" is the entire task, and the tag, date and formatting are somebody else's problem, specifically the AI's.
This guide covers what to say, how the words become a transaction, how to check and fix the result, which languages work, what needs a connection, and when the keypad is the better choice.
Why voice beats the keypad for everyday spending
A keypad entry is fast, three taps in a good app, but it still asks you to switch into data-entry mode: find the tag, decide on a note, confirm. Speaking skips the mode switch. You describe the purchase the way you would to a friend, and the description already contains everything a tracker needs.
There is also a quieter benefit. Studies on payment form, such as Raghubir and Srivastava's work on payment coupling, find that spending feels less real the further it gets from handing over cash. Saying the amount out loud, even to a phone, puts a small piece of that awareness back. It is hard to say "forty dollars on snacks" three days running without noticing.
In Cash Book, voice logging is the coral mic circle at the bottom of the Home screen. It is one of four ways to log, alongside the camera, an Apple Pay automation and the keypad; the wider system is in how to track expenses on iPhone.
What to say: amount, what, where
Three pieces of information make a complete entry, and the order does not matter.
- The amount. Say it the way you would naturally: "twelve fifty", "nine dollars", "four euros twenty". You do not have to name the currency if it is the one you set in Settings; the AI assumes it.
- What it was. "Lunch", "groceries", "a haircut", "the electricity bill". This becomes the note and is the main clue for the tag.
- Where. "At the deli", "at the market", "on the bus". The merchant becomes the note, and the tag is matched against your own tag names.
Optional extras that the AI also picks up:
- When. "Yesterday", "on Saturday", "this morning". The date moves accordingly. Without a time word, the transaction lands on today.
- Which account. "On the cash" or "from savings" if you keep several accounts. Otherwise the default account is used.
- Direction. "Ana paid me back twenty" or "got paid fifty for the desk" lands as income, not an expense.
You do not have to say the tag. The AI is given your own tag list and matches each item to the closest one, so the word "groceries" finds your groceries tag even if you called it Food shop.
Logging the whole day in one go
The single most useful voice habit is the end-of-day run-through. Hold the mic on the walk home, or while the kettle boils, and describe the day in one breath:
"Coffee three fifty this morning, lunch fourteen at the noodle bar, twenty on the bus card, eight at the pharmacy, and I got the ten back from Sam."
That is five transactions: four expenses and one income, each with its own amount, note, tag and date. One recording, under fifteen seconds. Same-day logging is the habit that keeps a tracker honest, and this is the cheapest way to keep it.
If you sometimes spend cash and sometimes tap a card, mention which when it matters. The AI can put the pharmacy on your cash account and the lunch on your card if you say so; how to track cash spending has more on keeping the cash side honest.
How the AI maps words to tags
Each recording is sent, with a little context, to an AI model through the app's relay. The context is what makes the result feel personal: today's date and weekday, your time zone, your currency, the names of your tags, your hashtags, your accounts and which one is the default. The model returns a list of transactions, and each one carries an amount, a currency, a merchant, a category or tag, a note, a date, and whether it is an expense, income or transfer.
Tag matching works on meaning, not exact words. "The gym fee" goes to a fitness or health tag if you have one. "Lidl forty-two" goes to groceries because that merchant is a grocer. "Drinks with Tom" goes to eating out or fun, depending on what your list offers. The AI does not invent tags; it chooses from yours. This is why a short, well-named tag list matters more for voice logging than for any other method, and why it is worth spending five minutes in tags and categories getting the names right.
When something does not fit, the model falls back to a sensible default and you change it with one tap. Over a few weeks you will notice which phrasings land cleanly and which need a nudge, and your sentences will adapt without you thinking about it.
Six example phrases and what they become
| You say | What lands in the app |
|---|---|
| "Twelve dollars, lunch, at the deli" | Expense, 12.00, note "Deli", tag Eating out, today |
| "Groceries forty-two at the market yesterday" | Expense, 42.00, note "Market", tag Groceries, yesterday's date |
| "Coffee three fifty and a croissant two" | Two expenses, 3.50 and 2.00, both tagged Coffee or Eating out, today |
| "Twenty on gas, from the cash" | Expense, 20.00, tag Transport or Fuel, on the cash account |
| "Sam paid me back the ten" | Income, 10.00, note "Sam", tag Paid back or your closest income tag |
| "Electricity bill eighty-six on Monday" | Expense, 86.00, tag Bills or Home, dated the most recent Monday |
The exact tag names depend on your list; the point is that the sentence needs no structure beyond what you would say to a person.
Review and edit before it sticks
Nothing is written silently. After the AI responds, the new transactions appear in the day's list on Home with an Edit option. Read them once. Most of the time they are right, and the review is a glance. When one is not, tap Edit and you get the same sheet the quick keypad uses: the amount, tag chips, note, date, account, and the Expense, Income or Transfer pill. Change what needs changing and save.
The two errors worth watching for are numbers that sound alike, "fifteen" and "fifty", and amounts that were meant for a different account. Both are two-second fixes in the moment and a puzzle a month later, which is one more reason to log the same day.
Languages and currencies
Voice logging understands spoken expenses in any major language, and you can mix languages in one sentence: a Spanish speaker naming a German supermarket gets the same result as an English speaker naming an American one. The amount is read in the currency you speak. If you say "forty euros" while your app currency is dollars, the model records the currency you named.
The interface itself ships in 26 languages, and Settings offers 128 currencies. For most people this means the tags, screens and voice entries all live in one language, which is what makes the tag matching feel effortless.
What needs a connection, and when to use the keypad
Voice logging needs a connection, because the parsing happens through the relay to the AI model. The same is true of snapping a receipt. Without a signal, the mic will not be able to return a result.
The keypad is the offline path, and it is always one tap away: the black plus button opens a big number, tag chips in the footer, and optional fields for note, date and account. Three taps for a typical entry. Use it when you are underground, on a plane, or in a shop with no signal, and the entries will look identical to voice entries afterward. It is also the better choice for one quick, precise number when you are in a quiet place and would rather not speak.
For anything on paper, a receipt, a bill, a screenshot of an order, the camera is the better tool. How to scan receipts into an expense tracker covers that path.
Privacy: what happens to the recording
The audio clip goes through the studio's relay to an AI model for parsing and is not kept after the transactions come back. The transactions themselves live on your phone and, if you turn on iCloud sync, in your own encrypted iCloud account. Usage analytics count that a voice log happened and whether it succeeded, never the words or the amounts. There is no sign-up and no account, so there is nothing to log in to and nothing to leak.
Making voice logging a habit
Three small habits turn voice logging from a feature into a routine.
- Speak on the way out. The moment after paying, while the amount is still on the screen or receipt in front of you, say it. It takes less time than putting the phone away.
- Keep a fixed catch-up point. Pick a moment, the walk home, the kettle, brushing teeth, and run through anything you missed. One recording covers the day.
- Read one number a week. Tap 1W on Home and look at the total. Then, once a month, open Insights to see which tag changed most. The voice entries are only useful if they feed something you look at.
Small amounts are the ones people skip, and they are also the ones that add up. The small purchases calculator shows what a daily "it's only three dollars" comes to over a year, which is a useful thing to know before deciding that coffees are not worth logging. Logging them by voice costs a few seconds and closes the gap that makes so many impulse purchases invisible.
What this is based on
The advice to log every purchase promptly and review weekly follows public money-management guidance from MoneyHelper and the Consumer Financial Protection Bureau, both of which begin budgeting with a complete record of actual spending. The Federal Reserve's Economic Well-Being of U.S. Households report describes how common it is for households to be surprised by their own spending. The link between how tangible a payment feels and how much people spend is from Raghubir and Srivastava's peer-reviewed research on payment form, listed below.
Get Cash Book on the App StoreFree for 7 days. $19.99 once for lifetime.
Frequently asked questions
What do I actually say to log an expense by voice?
The amount, what it was, and where, in any order that feels natural. “Nine fifty for a sandwich at Pret” or “Pret, sandwich, nine fifty” both work. Add “yesterday” or “on Monday” to move the date. You do not need to name the tag; the AI matches one from your list.
Can I log several purchases in one recording?
Yes. Run through the day as you would tell a friend: “coffee three, lunch twelve at the noodle bar, twenty on the bus card and eight at the pharmacy.” Each item becomes its own transaction with its own amount, note and tag. This is the fastest way to catch up on a day you did not log as you went.
Does voice logging work in my language?
It works in any major language, and you can mix in merchant names from another language without breaking it. The amount is read in the currency you speak, and the app itself ships in 26 interface languages, so the tags and screens match the language you think in.
How does it know which tag to use?
The AI is given your tag list and picks the closest match for each item. “Groceries at the market” goes to your groceries tag; “the gym fee” goes to fitness or health if you have such a tag. If nothing fits well, it falls back to a sensible default, which you can change with one tap.
Is my voice recording kept anywhere?
No. The clip goes through the app's relay to an AI model for parsing, and nothing is stored after the transactions come back. Usage analytics only count that a voice log happened and whether it succeeded, never what you said or how much you spent.
What happens if I have no signal?
Voice logging needs a connection, because the parsing happens through the relay. Without one, use the keypad: tap the black plus, type the amount, tap a tag, done. It takes three taps and works anywhere. The entries look identical afterward.
What if it hears the wrong amount?
Every voice log shows the result with an Edit option before you move on. Open the entry, correct the number, tag, note, date or account, and save. Nothing is written silently, so a misheard “fifteen” for “fifty” is a two-second fix rather than a mystery at the end of the month.
What this is based on
Get Cash Book on the App StoreFree for 7 days. $19.99 once for lifetime.