Put the phone on the table and talk. The transcript is written while the interview happens, the voices are separated when you stop, and your source's words never leave the phone.
To transcribe an interview on iPhone, record it in Pithflow. The transcript is written on the phone while you talk, and when you tap Stop the phone separates the voices into Speaker 1, Speaker 2 and so on, which you can rename. The audio never leaves the iPhone. The first 2 minutes of every recording are free; full-length interviews come with the Notes plan.
Most bad interview transcripts are bad recordings. The phone's microphones are at the bottom edge, so lay it flat between you and the other person with that edge pointing at them, on a notebook or a napkin rather than a bare table that carries every tap and cup. A quiet corner beats a busy café by more than any app setting can make up.
Two minutes the day before saves the interview: record a short test on Wi-Fi in the language you will use. That downloads the language's transcription pack and the speaker models once, so on the day the phone needs no connection at all — and you can put it in airplane mode, which also means no call can interrupt the recording.
After you tap Stop, Pithflow listens to the whole recording again on the phone and groups the turns by voice — the way a person can tell two speakers apart on a radio show without seeing them. Each group becomes Speaker 1, Speaker 2 and so on. Rename them once and the names carry through the transcript, the notes and every export.
It works best the way a good interview already sounds: people taking turns, each a similar distance from the phone. Two people talking over each other, or two voices that sound alike, can put a line under the wrong name — which is exactly what Change speaker is for (touch and hold the line). The first speaker-separated recording downloads the models once; after that, it runs offline. If a device cannot run speaker detection, the app says so and the transcript comes without labels.
Apple's built-in Voice Memos does not label speakers; its user guide describes a transcript you can view and copy, with no names. The one Apple feature that does label speakers is call recording in the Phone app, which only works on a phone call with one other person.
| For an interview | Voice Memos (built in) | Pithflow |
|---|---|---|
| Transcript while you record | Yes | Yes |
| Each speaker labeled | Not mentioned by Apple | Yes, renameable, fixable per line |
| Transcribe a recording made elsewhere | Not documented by Apple | Yes — import audio or video (Notes plan) |
| Get a file out | Copy the text only | Text, SRT subtitles, Markdown with the notes, or the audio (Notes plan) |
| Notes from the interview | A summary with Apple Intelligence | Summary, key points, decisions, action items |
| Cost | Free, iPhone 12 or later | 2 min of every recording free; $4.99/mo, $29.99/yr or $49.99 once |
Voice Memos per Apple's iPhone User Guide for iOS 27, read November 8, 2026.
Ask, and get the yes on the recording itself: "Is it OK if I record this so I quote you accurately?" takes five seconds and settles it. In the US, some states require everyone's consent and others only one person's; the details, and why it matters even with an app on the table, are in is it legal to record a meeting. Pithflow asks you to confirm that everyone agreed before every recording. This is not legal advice — your newsroom or school may have stricter rules than the law.
Yes, if it is a file the iPhone can open. In Pithflow's Notes library, tap Import audio, choose the language of the recording, then Choose a file. It takes audio or video from Files, Voice Memos or any app that can share a file; from a video it keeps only the sound. The file is transcribed and the speakers separated on the phone, like a recording made in the app. File import comes with the Notes plan.
A recording that lives in Voice Memos goes to Files first: in Voice Memos, tap the recording, tap More, tap Share, then Save to Files. The full walk-through is here.
Not a call on the same iPhone: iOS does not let any app record a phone call, and Pithflow tells you so if you try. If the call is on another phone or a computer on speaker, the iPhone can record the room. Apple's own Phone app can record a one-to-one call in some regions, with both people hearing a notice.
Yes, it separates as many voices as it can tell apart. It is most reliable when people take turns and sit at a similar distance from the phone. Where voices overlap, check the labels and use Change speaker on any line that landed under the wrong name.
No. Recording, transcription and speaker separation happen on the iPhone or iPad, and the audio stays there. iCloud sync, if you turn it on, carries transcripts and notes, not audio.
One language per recording, chosen before you tap Record from the ones your iPhone offers, English and Spanish among them. For an interview that switches languages mid-sentence, Pithflow Pro has a cloud mode that sends the audio with your permission.
The recording does not stop. On the free plan the first 2 minutes are transcribed and the rest of the audio is kept on the phone until you unlock it. With the Notes plan or Pro, every recording is transcribed in full.
Download Pithflow free and record a two-minute test before your next interview. No account, nothing uploaded.
