How to turn an Apple Voice Memo into text on your iPhone

The quick answer: two routes

There are two ways to get text out of a voice memo, and they solve different problems.

Route 1, Apple's built-in transcript. Free, already on your phone, works on the recording as it sits in Voice Memos. Good for short memos in one supported language. You get plain text and nothing around it.

Route 2, share the memo to a transcription app. Costs money, takes a few minutes, and gives you speaker labels, timestamps you can tap, a written summary, action items, and a PDF or Word file you can send.

If the audio you want is not a voice memo at all, a file sitting in Files, an attachment someone emailed you, a recording from a third-party app, the general routes are in how to transcribe audio on iPhone. This page stays on Apple's Voice Memos app.

Route 1: Apple's built-in transcript, step by step

You need an iPhone 12 or later, running iOS 18 or newer. The device line is the part most people miss. This is not Apple Intelligence, so you do not need the newest Pro model, but it does need an iPhone 12 or later: an iPhone 11, 11 Pro, 11 Pro Max or SE (2nd generation) will never show the transcript option, even after updating to the latest iOS. Apple also says transcription is not available in every country or region.

One thing before you start: Apple has moved this control between iOS versions. The current route is the three-dot More button on the recording, then View Transcript. On iOS 18 it often sat next to the playback controls as a small speech-bubble icon with lines in it, so if you see that icon, tap it and you land in the same place.

1. 1Open Voice Memos on your iPhone.
2. 2Tap the recording you want. The row expands into a player with a waveform.
3. 3Tap the three-dot More button, then tap View Transcript. On iOS 18 the same thing may sit next to the playback controls as a small speech-bubble icon with lines in it.
4. 4The text appears under the waveform. Press play and the current word highlights as it is spoken.
5. 5Tap any line in the transcript to jump the audio to that moment.
6. 6You can also watch the text arrive while you are still recording. During a recording, swipe up from the top of the waveform and the transcript scrolls live, with the current word highlighted as you say it.
7. 7To get the text out, tap More, then Copy Transcript for the whole thing. For part of it, open View Transcript, select the lines you want and tap Copy. Either way it is still copy and paste: paste into Notes, Mail, or a message.
8. 8To find an old memo, go back to the recordings list, pull down to reveal the search bar, and type a word from the recording. Voice Memos searches inside transcripts, not only the titles.
9. 9On iPad the layout is wider but the menu does the same job. On a Mac with Apple silicon running macOS 15 or later, Voice Memos transcribes recordings itself rather than borrowing the iPhone's text, and your recordings sync across through iCloud. An Intel Mac gets the audio and no transcript.

What Apple's transcript will not do

Everything below is a real limit, not a nitpick. For a two-minute note to yourself none of it matters. For an hour-long interview, most of it does.

• No speaker labels. A conversation between three people comes back as one wall of text with no idea who said what.
• No real summary and no action items. On an iPhone that supports Apple Intelligence you can select the transcript and run Writing Tools over it, which hands back a shorter block of prose. What you never get is minutes with owners, decisions and dates, or a list of tasks pulled out on its own. You get every word, including the throat-clearing, and you do the reading.
• You cannot edit it. If it hears a client's name wrong, the wrong spelling stays there forever, in every search result.
• No real export. Tapping Share on a voice memo sends the audio file, not the text. Copy and paste is the only way the words leave the app.
• No PDF, no Word file. If someone needs a document, you are building it yourself in another app.
• A short language list. Far shorter than most people expect. More on that next.
• One language per recording. A memo that mixes two languages comes back wrong, and there is no setting that fixes it.

The language limit is the one that catches people

Apple lists ten languages for Voice Memos transcripts: English (all variants), Spanish, Portuguese, Italian, French, German, Japanese, Korean, Simplified Chinese and Traditional Chinese. That is the whole list, and it moves between iOS releases. Apple also says transcription is not available in all countries or regions, so the right language on a phone set up in the wrong place can still come back with nothing. If your language is not on the list, the transcript option either does not appear or gives you nothing useful.

The fastest way to check your own phone is to record ten seconds in the language you care about and try to open the transcript. You will know in under a minute.

There is a second, harder problem. Even inside the supported list, a memo is treated as one language from start to finish. If you talk in Arabic and drop in English product names, or record in Hindi with English numbers and dates, the second language comes back spelled out phonetically or dropped. People notice this straight away, because the words that get mangled are exactly the ones that mattered: names, prices, deadlines.

Fixing that needs a transcription engine that tags language word by word instead of locking the whole file to one choice. Noter AI does this across 60+ languages with detection always on, so a memo that switches language mid-sentence keeps both halves intact. If you record in two languages regularly, the fuller comparison is in the best app for transcribing multilingual meetings.

Route 2: send the memo to a transcription app

This is the route when you need more than the words. It uses the normal iOS share sheet, so it works with the recording exactly where it already is. No exporting, no cables, no computer.

The share sheet works the same for any audio on the phone, and the full walkthrough, including files that never went near Voice Memos, lives in how to transcribe audio on iPhone. Here is the short version for a recording already sitting in Voice Memos.

1. 1Open Voice Memos, tap the recording, tap the three-dot More button, then tap Share.
2. 2Pick the transcription app in the share sheet. If it is not showing, swipe left along the row of app icons, tap More, switch the app on and drag it near the top.
3. 3The audio transfers and transcription starts on its own. A normal recording takes a few minutes, longer files take proportionally longer.
4. 4Read the result: the full transcript, tappable timestamps, speaker labels where there is more than one voice, and a summary above it. Export what you need as PDF, Word (.docx), plain text, or a rich-text copy.

Voice Memos transcript vs a full transcription app

One caveat on the summary row. On an iPhone that supports Apple Intelligence you may see a short generated line above the transcript, and you can also select the transcript and summarise it with Writing Tools. Both give you shorter prose about the recording, not minutes with decisions, owners and dates.

The last row is the honest trade. Apple's transcript never leaves your phone. A cloud tool uploads the audio, does more with it, and sends the result back. If your rule is that the audio must stay on the device, Route 1 is the only route, and the limits above are the price.

Clearing out a backlog of old voice memos

Most people who search for this have a list of 200 recordings called New Recording 47 and no idea what is in them. The built-in transcript is genuinely useful here, because you can identify a recording in five seconds without listening to it.

1. 1Open the oldest recording and open its transcript. Read two lines. That is usually enough to know what it is.
2. 2Rename it to something you would actually type into a search box: Rent call, landlord, June beats New Recording 47.
3. 3Delete it if it was a one-off reminder. Most of the backlog will be this, and deleting is faster than organising.
4. 4For the ones worth keeping, share them into a transcription app one at a time, so they come back with a summary and searchable text.
5. 5Put the imported notes into folders by client, project or month.
6. 6Search across everything from then on. In Noter AI you can also ask a question across all your notes at once, for example what you agreed to send someone in June, and get the answer with the source note attached.
7. 7Once a memo is imported and exported to PDF, delete the original if you are short on storage. Long recordings are large files.

Doing the same thing on Android

Android has no Voice Memos app, so the answer depends on the phone.

Pixel. The Recorder app records and transcribes on the device, in real time, and lets you search inside recordings. It works with no signal. Language support is narrower than the marketing suggests, and speaker separation is limited.

Samsung Galaxy. Voice Recorder has speech-to-text, and newer models with Galaxy AI add transcripts and short summaries. Which features you get depends on the model and the language.

Everything else. Most stock recorder apps only record. You get an audio file and no text at all.

The share route works the same way on any Android phone.

1. 1Open your recorder app and find the recording.
2. 2Long-press it, or open it and tap Share.
3. 3Pick the transcription app in the Android share sheet.
4. 4Transcription starts, and the finished note has the transcript, speaker labels, timestamps and summary.
5. 5Export to PDF, Word or plain text from inside the app.

Troubleshooting

Five things go wrong more often than everything else combined. Here is what each one actually is, and what fixes it.

• There is no transcript button anywhere. Three causes, in the order worth checking. First the phone: transcription needs an iPhone 12 or later, so an iPhone 11, 11 Pro, 11 Pro Max or SE (2nd generation) never gets it, whatever iOS version it is running, and nothing fixes that one. Second the software: you need iOS 18 or newer, so look in Settings, General, Software Update. Third the language or the region: the recording has to be in one of Apple's ten supported languages, and Apple says transcription is not available in all countries or regions. If the phone and the iOS version are fine, it is one of those last two, and no setting will change it.
• The transcript is empty, or stops after a few words. The recording is too quiet or too noisy for the words to be picked out. A phone in a bag or a pocket, a hand over the bottom edge, an air conditioner, or a speaker sitting far across a room will all do it. You cannot repair a bad recording afterwards. Next time put the phone face up on the table with the bottom edge pointing at whoever is talking.
• Names and technical terms come out wrong. Apple's transcript is read-only, so your only option is to copy the text into Notes and fix it there, which means the search inside Voice Memos still has the wrong spelling. Tools with an editable transcript let you correct the word once and then rebuild the summary from the fix, explained in editing an AI transcript and re-running the summary.
• Half the memo reads like nonsense. This is the mixed-language failure. The engine committed to one language and wrote the other one out by sound. Nothing in Voice Memos fixes it. Run the same audio through a tool that detects language per word and it comes back readable.
• The app you want is missing from the share sheet. Swipe left to the end of the icon row, tap More, switch the app on, and drag it to the top. Make sure it is installed and has been opened at least once, since a share extension does not register until then.
• The share or import fails on a long recording. A one-hour memo is a big file. Check you have free storage, and that you are on Wi-Fi if the app processes in the cloud, since the whole file has to upload.
• The memo is greyed out or will not open. It is stored in iCloud and not downloaded to this phone yet. Tap it once and wait for it to come down before sharing it anywhere.

If you need more than the words

Apple's transcript answers one question: what was said. Everything else you would want from a recording, it leaves to you.

Noter AI takes the same voice memo through the share sheet and gives back the parts Voice Memos skips. Speaker labels, so a three-person conversation reads properly. Timestamps on every segment, with synced playback, so tapping a line plays that exact moment. Action items pulled out on their own. A summary in whichever shape you need: minutes of meeting, key bullet points, an email, a to-do list, a formal report, or a custom instruction you write yourself, describing exactly how you want it. The transcript is editable, and correcting a name lets you re-run the summary so the fix carries through.

It also handles the two things this page keeps coming back to. It transcribes in 60+ languages with automatic detection, tagging language word by word, so bilingual recordings survive. And it translates a finished note into 18 languages: Arabic, Chinese (Simplified), Chinese (Traditional), Dutch, English, French, German, Indonesian, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish, Thai, Turkish and Vietnamese.

It costs $9.99 a month, or $49.99 if you pay for the year, on a flat plan rather than a seat licence. Voice memos are the awkward case for metered plans, because you do not decide in advance how many you will record. A week of site visits or a run of long dictations can empty a monthly minute or credit allowance without you noticing until a transcript refuses to generate. Nothing here is rationed. For reference on the per-user alternatives: Otter.ai Pro is $16.99/month, Notta Pro $13.99, Granola $14 on Business, tl;dv Pro $18. There is a free trial, so test it on a real recording before paying.

The honest limits: transcription happens in the cloud, not on the device, so it needs a connection and the audio does leave your phone. It takes audio files only. And there is no live caption while you are still recording, the transcript arrives after you stop, which is the one thing Apple's own Voice Memos does do.

It runs on iPhone and iPad and on Android, and notes sync across your devices. If you also record meetings rather than only memos, the wider picture is in the best AI meeting note taker for iPhone.