How to transcribe audio to text for free, right from your phone

What does free transcription actually include?

Free transcription is real, but it is rarely the same offer twice. Before you trust a two-hour recording to a tool, check four things.

• Length per file. Many free tiers stop after a few minutes — fine for a voice note, useless for an interview.
• Volume. Some tools give you a set amount of audio to start with, others reset a counter each month.
• What you can take away. The text may be free to read but locked behind a plan for PDF or Word export.
• What you have to give. An email address, a card to open a trial, or ads sitting between you and your transcript.

None of that makes a tool bad. It simply decides whether the tool fits the file in your hand today. Read the limits before you record, not after.

Why your phone is the fastest free route

Most audio already lives on a phone: a voice memo, a WhatsApp message, a meeting recorded in a pocket, a file someone dropped in Drive. Moving it to a computer to use a web uploader often costs more time than the transcription itself.

On both iPhone and Android, the share sheet does the work. Open the app holding the audio, tap Share, choose your transcription app, and the file goes straight in — nothing to download, rename or convert. Zeepit works this way: pick a file from Files or Google Drive inside the app, or share any audio to it from another app. See how the Share button works.

The other free option is live dictation, built into both systems. It types while you speak, which is handy for a quick note, but it cannot read a file you already have, and it usually returns one long block of text with no summary.

Step by step

1. Find the audio on your phone

   A voice memo, a WhatsApp message, a meeting recording, a file in Files or Google Drive. You do not need to convert it or move it to a computer first.

2. Send it to a transcription app

   Tap Share on the audio and choose the app, or open the app and pick the file from your storage. The gesture is the same on iPhone and Android.

3. Let the language be detected

   Good tools detect the spoken language on their own, so Portuguese, German, Japanese or Turkish audio needs no setting. Start the conversion and leave the screen if the file is long.

4. Read the transcript and fix the names

   Errors cluster around names, brands and acronyms. Scan those first, correct them, then read the rest. In Zeepit, copying the text or the summary takes one tap and is free.

5. Summarise, translate or export

   Turn the transcript into a short, medium or long summary, translate it if the audio was not in your language, then copy it or export it as PDF, Word or plain text.

Which audio and video files can be transcribed?

Free tools refuse files for two boring reasons: the format and the length.

The common voice formats are mp3, m4a (iPhone voice memos), wav, caf, ogg, opus (WhatsApp and Telegram), aac, flac, amr and webm. Video counts too — mp4, mov, mkv and avi files can be transcribed, since only the audio track is read. Zeepit accepts all of those and takes recordings of up to four hours each, so you rarely need a converter at all.

If a tool rejects your file, do not re-record the meeting. Re-encoding audio a second time adds artefacts and can make the transcript worse. Try another app first, and keep the original file rather than a copy that has been compressed by passing through several messaging apps.

Recording habits that cut your editing time

No tool, free or paid, recovers words it never heard. Most transcript errors come from the recording, not from the software.

• Get the microphone close. A phone on the table beats a phone in a pocket; one metre beats three.
• Remove the constant noise. Air conditioning, a fan, a café playlist — steady background sound hurts accuracy more than the occasional cough.
• One voice at a time. Overlapping speakers are the hardest case for every system.
• Say the hard words once, clearly. Names, brands and acronyms spoken at the start give you something to search and replace afterwards.
• Keep the original file. Forwarding audio through several apps compresses it each time.

Two minutes of setup usually saves more time than any post-editing trick.

Do you need the full transcript, or just a summary?

Word-for-word text is what you want for an interview you will quote, a lecture you will study, or anything that may be checked later. For most other audio, you only want the point.

A good tool gives you both from the same file: the complete transcript, and a summary you can set to short, medium or long depending on whether you need three lines or a full page of notes. Zeepit also takes a YouTube, podcast or web page link and returns a summary, which is a quick way to decide whether a two-hour episode deserves your evening.

Voice notes are the clearest case: a rambling three-minute message becomes two lines you can act on. If most of your audio comes from messaging, see our guide to WhatsApp voice notes.

What to check before you send a recording anywhere

Audio is personal: a medical appointment, a client call, a message from a relative. Before sending a file to any service, ask three questions.

• Where does the file go? Almost every accurate tool processes audio on a server. That is normal; what matters is what happens next.
• How long is it kept? Look for a clear statement that the file is deleted after processing rather than archived.
• Is it reused? Check whether your recordings feed anything beyond your own transcript.

Zeepit's answer: the file is sent, transcribed and deleted right after — no archive, no reuse, no AI training on your recordings. The transcript and the summary stay on your device, where copying them takes one tap and costs nothing. Export to PDF, Word or plain text is generated on the phone and is part of Zeepit Pro.

Frequently asked questions