How do I get a transcript with timestamps on iPhone?

A transcript without timestamps is a wall of text you cannot verify. Timestamps turn it into an index: you read a line, doubt it, and jump straight to that second of audio. Most free browser tools return plain text; the ones that keep timing information are where the useful work happens. AI Transcribe applies timestamps and speaker labels automatically on iPhone, in any of 32 languages, with the language detected for you.

What timestamps are actually for

Three jobs, in order of how often they matter:

1. Verification. Any claim that carries consequences — a quote, a number, a commitment — should be checkable against the audio. The timestamp is the check.
2. Navigation. In a two-hour recording, finding "the bit about the budget" without timing information means scrubbing blindly.
3. Citation. Journalists, researchers and anyone writing from interviews need to point at when, not just that.

Paired with speaker labels they become considerably more useful: "who said it, and when" is the pair that makes a transcript evidence rather than notes. Speaker labelling is its own problem — see what speaker diarization is for how it works and where it fails.

Automatic language detection, and where it trips

You do not need to tell the app what language a recording is in — detection handles it. Two situations to know about:

• Code-switching. When a speaker moves between languages mid-sentence, detection picks the dominant one and the minority language transcribes poorly. If you know a recording is genuinely bilingual, expect to clean it up.
• Very short clips. Detection has less to work with in a ten-second voice note than in an hour of speech. Short clips in less-common languages are the weakest case.

Neither of these is specific to one app — they are properties of how detection works.

Getting both on a phone

1. Record in the app (up to two hours) or import an existing audio or video file (up to three hours).
2. The transcript comes back with speaker labels and timestamps applied — nothing to configure.
3. Use them: tap to jump, or ask questions against the transcript and check the answers at the timestamps they came from.

Free trial, which is enough to check the timestamp accuracy on your own audio before paying for anything.

The honest constraints

Timestamp precision follows transcript quality. In a clean recording they land where you expect; in a noisy one with overlapping speech, both the labels and the timing drift — and that drift is a signal that the recording, not the app, is the limiting factor. Transcription runs in the cloud, so a connection is required.

Frequently asked questions

Do I have to turn timestamps on? No — they are applied automatically along with speaker labels.

Do free transcription tools include timestamps? Often not. Plain-text output is the common free-tier shape; timing and speaker information are usually where the paid line sits.

Does it detect the language automatically? Yes, across 32 languages. You can expect trouble only with heavy code-switching or very short clips.

Can I get timestamps for an imported file, or only live recordings? Both. Imported audio and video are treated the same as recordings made in the app.