Turn any voice memo to text, free, in your browser

How to convert a voice memo to text on iPhone

Apple does not expose Voice Memos as a normal folder, so the only real work is getting the file out of the app. After that it is an ordinary upload — from the same iPhone if you like.

1. Open the Voice Memos app and tap the recording you want.
2. Tap the three-dot menu on the recording, then choose Share.
3. Pick Save to Files and drop it in iCloud Drive or On My iPhone. AirDrop to a Mac or emailing it to yourself works just as well.
4. Open nemotronasr.com in any browser — Safari on the same phone is fine — and tap the upload area at the top of this page.
5. Select the memo from Files. It is an M4A file, which the transcriber reads directly.
6. Leave the language on auto-detect or pick one of the 40 locales, transcribe, then copy the text or download it as a .txt file.

Voice memo to text on Android

Android is simpler because recorder apps save ordinary files you can reach from any file manager. Samsung Voice Recorder and Google Recorder on Pixel both write M4A. Most third-party recorders use MP3, WAV, AAC, or OGG. Every one of those formats is accepted here.

Open this page in Chrome, tap upload, and pick the recording from Files, Downloads, or Drive. Nothing to export first, nothing to convert. The same goes for voice notes shared from chat apps: an exported WhatsApp voice note is an OPUS file, and OPUS uploads are supported too.

Why this works: voice memos are ordinary audio files

The free tier accepts MP3, WAV, M4A, AAC, OGG, and OPUS files up to 5 minutes and 20 MB. Every mainstream voice recorder lands inside that list, which is why this page has no converter step: the transcriber is the converter.

iPhone already transcribes memos on iOS 18 — so why use this?

Fair question, and worth answering honestly. As of 2026, recent iPhones on iOS 18 or later can show a transcript inside the Voice Memos app itself. If you have one of those phones, you recorded in English, and the built-in transcript reads well — use it. It is free and already on your device.

This tool earns its keep everywhere the built-in feature stops:

• Languages. Apple's memo transcripts cover a short list of languages. Nemotron handles 40 locales — Spanish, Mandarin Chinese, Japanese, Korean, Hindi, Arabic, and thirty-plus more — with auto-detection that keeps up even if you switch languages mid-memo.
• Older iPhones and older iOS versions that never got a transcript button.
• Android. Built-in transcription on Android depends on the maker and model — some recorder apps have it, many do not, and language coverage varies. This page works the same on every Android phone.
• Desktop. Drag a memo into a browser tab on Mac, Windows, or Linux — no phone in hand required.
• Nothing to install, update, or grant permissions to. It is a web page.

What a transcribed memo is actually good for

One caution on meeting recordings: the output is plain punctuated text, not a speaker-labelled script. Nothing marks who said what. If you need per-speaker labels — diarization — use a dedicated meeting tool; this one will not pretend to do it.

Four speed modes let you trade accuracy against wait time. Best accuracy suits memos recorded in a car or on the street; Lowest latency is for when you just need the gist as fast as possible. Balanced is the sensible default.

• Ideas you dictated while walking or driving — turn a rambling minute of audio into a note you can search and edit.
• Personal notes and voice journals — a text copy is far easier to skim than a folder of thirty recordings.
• Reminders and to-dos — transcribe, then paste straight into your task app.
• Short interviews and vox pops — one speaker at a time gives the cleanest result.

Free limits, stated plainly

• Free: 3 transcriptions a day per connection, each file up to 5 minutes and 20 MB. No account, no card, no email.
• A typical voice memo — an idea, a reminder, a note-to-self — runs well under 5 minutes, so the free tier covers normal memo use comfortably.
• Longer recordings, like lectures or full interviews, exceed the free cap. Pro is $9 per month for 300 audio minutes, activated with a Creem license key in up to 3 browsers.

What this tool will not do

Knowing the edges saves you time. These are hard limits, not settings you have missed:

• No speaker labels and no timestamps. Output is plain punctuated, capitalized text you copy or download as .txt.
• No SRT or VTT export — subtitles need per-segment timing, which this tool does not produce.
• No video files. Extract the audio track first, then upload that.
• No batch uploads and no public API. One file at a time.
• Transcription runs after a recording stops. It is not a live caption overlay.