Turn a recording, an imported file or live dictation into accurate text on your iPhone, iPad or Mac - with speakers labelled, timestamps you can jump to, and nothing uploaded.
No. Recognition runs on your iPhone, iPad or Mac. There is no account and no upload step, and the app works with the network switched off entirely - which is also the simplest way to prove to yourself that nothing is leaving.
m4a, mp3, wav, aiff and caf for audio, and mp4 and mov for video. Importing exists because the recording you most need a transcript of is usually one somebody else made and sent you.
Press Control-Option-D and dictation starts wherever the cursor already is, in whatever app you are using. Text lands directly in that document instead of in a separate window you then copy from.
It separates the voices in a recording and lets you name them, which is what makes a transcript quotable. It works best when speakers do not talk over each other and the microphone is roughly equidistant - a phone in the middle of a table beats a phone in front of one person.
Yes. Export as SRT or WebVTT and the timings come from the same pass that produced the transcript, so the captions line up with the audio they were made from.
Plain text for pasting somewhere quickly, Markdown when you want speaker names and structure preserved in a notes app or document, SRT or WebVTT for video captions, and JSON when another tool is going to read the transcript rather than a person.
Real-time transcription needs a newer system; on older ones the text arrives after the recording rather than during it. The result is the same either way - only the timing of when you see it differs.
Well over ninety depending on the engine, including German, Spanish, French, Italian, Portuguese, Dutch, Polish, Czech, Ukrainian, Swedish, Danish, Finnish, Greek, Turkish, Arabic, Hebrew, Hindi, Chinese, Japanese and Korean.
Yes. Every transcript goes into a searchable library, and selecting a line moves the audio to that timestamp - which is the normal way to check a quote before you use it.
iPhone on iOS 17, iPad on iPadOS 17, or a Mac on macOS 14, and later versions of each.
Record, import or dictate and get accurate text in seconds - entirely on device, exporting to notes, documents or subtitles.
The decisions that matter are made before the meeting starts, not afterwards.
Tell the room it is being recorded. That is a courtesy in most places and a legal requirement in some.
Try this in Voice to Text: Offline AI Note →Works the same for an audio file and for the audio track of a video.
If the source is a video you are going to publish, export SRT rather than text - you get the transcript and the captions from one pass.
Try this in Voice to Text: Offline AI Note →This is the mode people underuse, because it does not require opening the app at all.
Dictation is the fastest route from a thought to a written line, and it is worth learning the shortcut properly - it removes the step where you open an app first and lose the sentence you had.
Try this in Voice to Text: Offline AI Note →The aim is captions whose timings came from the real audio rather than being nudged by hand.
Proper nouns are the reliable failure case in every transcription engine. Reading for names first, before anything else, catches most of what a viewer would notice.
Try this in Voice to Text: Offline AI Note →Our support team usually replies within one business day.
Download Voice to Text: Offline AI Note Contact support