How to Transcribe iPhone Voice Memos: Free Built-in + AI Alternatives (2026)
iOS 18 added a free on-device transcription button to the Voice Memos app on iPhone 12 and later. This guide shows you how to use it, when it works well, and how to handle the cases where you need speaker labels, more languages, or AI summaries.
Voice Memo formats accepted by VexaScribe:
TL;DR
iOS 18 added a free built-in transcription button to the Voice Memos app on iPhone 12 and later. It works well for standard recordings in 10 supported languages, runs on-device (private), and requires no signup.
Use Apple's built-in feature first — it's genuinely good enough for most single-speaker recordings.
Upgrade to a third-party tool like VexaScribe if you need multi-speaker diarization, non-Apple languages (99 vs 10), long recordings (2 hours+), AI summaries with action items, SRT export for video subtitles, or shared team access.
Method 1: Apple's Built-in Voice Memos Transcription
Recommended for most single-speaker recordings in a supported language. Free, private, no signup.
Requirements
- ✓iPhone 12 or later (12, 13, 14, 15, 16, and newer)
- ✓iOS 18 or newer
- ✓Recording in one of 10 supported languages (see full list below)
Step-by-step
- 1
Open the Voice Memos app
Find the Voice Memos app on your iPhone home screen or App Library — it's the red icon shaped like an audio waveform.
- 2
Tap the recording you want to transcribe
Any existing recording will work. If you haven't recorded yet, tap the red record button, capture your audio, then tap Done.
- 3
Tap the transcript button
The transcript button (an icon showing lines of text) sits below the playback controls. If you don't see it, make sure you're on iOS 18 or newer.
- 4
Wait a few seconds
The transcript appears inline. Short recordings transcribe almost instantly; longer ones take a few seconds while the on-device model processes.
- 5
Tap any timestamp to jump to that moment
Words are synced to audio playback. Tap a sentence to hear it — perfect for verifying accuracy or finding a specific quote.
- 6
Share the transcript as plain text
Tap the Share button, then choose Messages, Mail, Notes, or any app that accepts text. Note: Apple exports plain text only — no DOCX, PDF, or SRT.
Supported languages (10 total)
Note: eight English variants share the same underlying model but are region-specific. For any other language (Turkish, Arabic, Hindi, Polish, etc.), Apple's built-in won't transcribe — see Method 2.
Privacy
On-device processing — audio never leaves your iPhone.
Cost
Free. No account, no subscription, no per-minute fees.
Limits
No speaker labels, older iPhones excluded, 10 languages only, no SRT export.
What Apple's built-in doesn't do:
- ⚠No speaker labels (single flowing transcript)
- ⚠Older iPhones excluded (iPhone 11 or below)
- ⚠Non-supported languages excluded
- ⚠No AI summary or action-item extraction
- ⚠No SRT/VTT/DOCX/PDF export — plain text only
Source: Apple Support — View a transcription in Voice Memos on iPhone.
When Apple's Built-in Falls Short
Apple's built-in transcription is genuinely good — but it wasn't designed for every use case. Here's where you'll hit walls.
Multi-speaker meetings or interviews
Apple gives you a single flowing text with no speaker labels. You'll spend more time formatting than transcribing.
Non-supported language
Turkish, Polish, Arabic, Vietnamese, Thai, Hindi, Russian, Dutch, and 80+ more — Apple's built-in simply won't attempt it.
Long recordings (2 hours+)
Apple works but processing can be slow on-device, and there's no progress indicator for very long files.
Poor audio quality
Background noise, distant speaker, phone mic clipped by clothing — Apple degrades noticeably faster than cloud models like Whisper Large-v3.
Need AI-generated summary with action items
Apple gives you the transcript and nothing else. No key-points extraction, no action items, no topic breakdown.
Need SRT/VTT export for video subtitles
Apple only produces plain text. If you're captioning a video (YouTube, Instagram, TikTok), you need timestamped SRT/VTT.
Team collaboration or shared access
Apple's transcript lives inside your Voice Memos app. Sharing means copy-paste. No shared workspace, no comment threads.
Older iPhone (iPhone 11 or earlier)
Feature not available. iPhone 12 was the cutoff. If you're on an older device, export and upload is your only path.
Non-English content mixed with English
Apple assumes one language per recording. Bilingual meetings and code-switching get transcribed poorly.
If any of these apply to your recording, skip to Method 2 below.
Method 2: Export Voice Memo + Transcribe with VexaScribe
Two-step workflow: export the M4A off your iPhone, then upload to VexaScribe for speaker labels, 99-language support, AI summary, and full-format export.
Step 1: Export the Voice Memo from your iPhone
Four options — pick whichever fits your workflow.
Option A: AirDrop to Mac (fastest, highest quality)
Open the Voice Memo, tap Share, choose AirDrop, then select your Mac. The M4A file arrives in your Mac's Downloads folder in seconds. No compression, no quality loss.
Option B: Save to Files (iCloud or local)
Share → Save to Files → choose iCloud Drive (syncs across all devices) or On My iPhone (stays local). Great if you want to keep the M4A in your file system for archival.
Option C: Mail to yourself
Share → Mail → enter your own email address → send. Download the M4A attachment on any device. Simple, works everywhere, no cloud storage needed.
Option D: Cloud share (Drive, Dropbox, OneDrive)
Share → pick any installed cloud app. The M4A uploads directly. Best if you're already living in Google Drive or Dropbox for collaboration.
File format note: Voice Memos records as M4A (AAC codec) by default. VexaScribe accepts M4A directly — no conversion needed.
Step 2: Upload to VexaScribe and transcribe
Same six-step flow whether you're on iPhone Safari, iPad, Mac, or PC.
- 1
Open vexascribe.com/dashboard
In any browser — iOS Safari on iPhone works fine if you'd rather stay on your phone. Chrome, Firefox, Edge, or Safari on Mac all work identically.
- 2
Sign in or create a free account
New accounts get 30 minutes free at signup — no credit card required. Enough to test with a full-length lecture or interview.
- 3
Drag or tap Upload
Select the M4A from Files, Photos (if AirDropped as attachment), or your downloaded folder. VexaScribe accepts M4A directly — no conversion needed.
- 4
Choose language and enable speaker detection
Auto-detect handles most cases; specify manually for accented or mixed-language recordings. Toggle speaker detection on if multiple speakers are in the recording.
- 5
Wait 5-10 minutes for a 1-hour Voice Memo
Processing scales roughly with duration. A 15-minute memo finishes in 2-3 minutes; a 2-hour lecture takes 15-20 minutes.
- 6
Export or view in the editor
Export as TXT, DOCX, SRT, VTT, or PDF. Or open the built-in editor to rename speakers, edit misheard words, and ask AI Chat questions about the recording.
Cost and accuracy
Cost
- • 30 min free at signup (no credit card)
- • $2/mo — 200 min ($0.60/hour effective)
- • $5/mo — 1,000 min
- • $10/mo — 2,500 min
- • $20/mo — 6,000 min
Accuracy
- • ~92-95% for clean English (Whisper Large-v3)
- • Drops with noise, heavy accents, technical vocab
- • Automatic diarization: Speaker 1, Speaker 2, etc.
- • Rename speakers after transcription
- • Optional AI summary with action items + topics
For a deeper look at model accuracy across accents and audio quality, see our Whisper accuracy deep-dive.
Common Use Cases with iPhone Voice Memos
Six real scenarios — and which method fits each.
Journalist interview
Speaker-labeled transcript with timestamps for article quotes. Apple built-in gives you the text; VexaScribe gives you the speaker labels.
Sales call
Transcript + AI summary of decisions and next steps. Voice Memo → export → VexaScribe → paste into SFDC or HubSpot.
Doctor's appointment
Transcript for personal records. Apple built-in is perfect for this — single speaker, private, stays on your phone.
Lecture recording
Study notes with chapter markers. VexaScribe's AI summary structures the lecture into topics you can revisit.
Meeting note-taking
Structured summary for team share. Speaker labels + action-item extraction save you the manual formatting pass.
Podcast recording
SRT export for the YouTube upload of your podcast video. Apple built-in doesn't produce subtitle files.
Voice Memos Quality Tips
Small habits that meaningfully improve transcription accuracy — for Apple's built-in and VexaScribe alike.
Use the built-in mic close to the speaker
3-6 inches from mouth for best quality. The iPhone's mic is directional — distance kills accuracy fast.
Reduce background noise before recording
Close windows, silence notifications, mute other devices. Silent room > noise-cancelling software after the fact.
Use an external mic for interviews
A Rode Wireless Go or Shure MV88 dramatically improves accuracy — often the difference between 88% and 96%.
Set audio quality to Lossless in Settings
Settings → Voice Memos → Audio Quality → Lossless. Uses more storage but preserves detail for transcription.
Speak clearly and pause between speakers
For interviews, a half-second gap between speakers helps diarization tools separate voices cleanly.
Test with a 30-second recording first
Before a long important session (lecture, deposition, keynote), record 30 seconds and check playback. Fix mic placement now, not later.
iPhone Voice Memo Transcription FAQ
Does the iPhone Voice Memos transcription work offline?
Yes. Apple's built-in Voice Memos transcription runs entirely on-device on iPhone 12 and later — audio never leaves your phone.
Which iPhones support Voice Memos transcription?
iPhone 12 and later, running iOS 18 or newer. Older iPhones (11 or earlier) don't have this feature.
Does Apple's built-in support Turkish (or Arabic, Hindi, Polish, Vietnamese)?
No. Apple's built-in supports 10 languages: English (multiple variants), Spanish, Portuguese, Italian, French, German, Japanese, Korean, and both Chinese variants. For any other language, use a third-party tool that supports 99+ languages via Whisper.
Can I transcribe multiple speakers?
Apple's built-in doesn't label speakers — you get a single flowing transcript. For journalist interviews, sales calls, or podcast conversations where you need to know who said what, use a tool with speaker diarization like VexaScribe.
How do I export a Voice Memo to my computer?
Open the Voice Memo, tap the Share button (square with up arrow), then choose AirDrop to Mac, Save to Files, Mail (send to yourself), or any cloud service (Drive, Dropbox, iCloud).
Is there a cost for Apple's built-in transcription?
No. Apple's built-in Voice Memos transcription is completely free. Third-party tools charge per minute or per month for advanced features like diarization, multi-language support, and AI summaries.
Related Guides
Voice Memo to Text
General voice memo transcription — iPhone, Android, and other recording apps.
M4A to Text
The format Voice Memos produces — direct conversion guide for M4A audio files.
Transcribe Audio
General audio transcription for any format — the hub page for all audio-to-text workflows.
Speaker Identification
Automatic speaker diarization for interviews, meetings, and multi-voice recordings.
How to Transcribe an Interview
Step-by-step guide with speaker labels and timestamps — extends the Voice Memo workflow to interviews.
Audio to Notes
Convert Voice Memos into structured lecture or meeting notes with AI summary.
Multilingual Transcription
99+ language support — for the 80+ languages Apple's built-in doesn't cover.
How Accurate Is Whisper
Deep-dive on the model powering VexaScribe — accuracy benchmarks vs. Apple built-in.
Try VexaScribe free — 30 minutes at signup, no credit card
Get speaker labels, 99+ languages, AI summaries, and SRT/DOCX/PDF export for the Voice Memos Apple's built-in can't handle.
Start freeFor simple English single-speaker Voice Memos, Apple's built-in feature works well — no signup needed. Just open Voice Memos and tap the transcript button.