Sbobinare: how to turn a recorded lecture into readable text

What does sbobinare mean, and what do you get at the end?

In Italian, sbobinare comes from bobina, the tape reel: it meant winding the tape back and typing out what was on it. The word survived the tape. Students use it for lectures, journalists for interviews, and anyone sitting on a two-hour meeting recording is doing the same job.

What you end up with is two different documents, and they are not interchangeable:

• The transcript: everything that was said, word for word. Long, repetitive, ugly to read, and searchable. It is where you find the exact sentence in which the lecturer said what would be on the exam.
• The summary: the same content reduced to what matters, in a form you can revise from.

Keep the transcript as the archive and work from the summary. Producing only one of the two is the most common mistake.

Recording well is half the job

No tool recovers what the microphone never captured. The five minutes before the lecture matter more than any option you pick afterwards.

• Lay the phone flat on the desk, screen up, bottom edge pointing at the speaker: that is where the main microphone usually sits.
• Sit in the front third of the room. Distance costs more clarity than background noise does.
• Never record from a bag or a coat pocket. Fabric kills the frequencies that carry speech.
• Silence the phone. A vibration against a wooden desk is louder in the file than in the room.
• Check storage and battery. A three-hour recording that dies at minute forty is the classic disaster.
• Record five seconds and play them back before you rely on the setup.

If the hall has a PA system, sit near a loudspeaker rather than near the lecturer. And where the rules require it, ask permission before pressing record.

Step by step

1. Record close to the speaker

   Put the phone flat on the desk near the lecturer, screen up and silenced, with enough battery and storage for the whole session. Play back five seconds before you rely on it.

2. Send the file to Zeepit

   Open Zeepit and pick the recording from Files or Google Drive, or tap Share on the audio in your recorder, WhatsApp or Telegram and choose Zeepit. Same gesture on iPhone and Android.

3. Let it transcribe, then check the language

   Wait for the transcript to appear. The language is detected automatically; on a multilingual course, confirm it matches the lecture before you work on the text.

4. Fix names and terms, then add headings

   Skim the transcript for wrong proper nouns, acronyms and technical words and correct them. Insert a heading each time the topic changes so you can navigate the file later.

5. Generate the summary and keep it

   Choose the summary length you need, then copy it with one tap, or export the transcript as PDF, Word or plain text so it lives somewhere other than the phone.

How do you sbobinare an audio file straight from your phone?

You do not need a computer. The recording is already on the phone, so that is where it can become text.

Two ways in. Either you open the app and pick the file from Files, Google Drive or wherever it sits, or you start from the audio itself: tap Share on the voice memo, the WhatsApp note or the Telegram file and choose the transcription app in the share sheet. This is how the Share button works, and it is the same gesture on iPhone and Android, with no upload, no format conversion and no download.

Zeepit takes mp3, m4a, wav, ogg, opus, aac, flac, amr, caf and webm, plus video files (mp4, mov, mkv, avi) where only the audio track is used. One recording can run up to four hours. The first conversions after installing are free, long recordings included, with no account, no card and no ads.

For voice notes in particular, the WhatsApp voice-notes guide covers the same route step by step.

What does sbobinare audio gratis actually get you?

Three honest options, with what each really costs.

• Typing it out yourself. Free and accurate, but it takes several passes over the same audio. Sensible for a ten-minute interview, brutal for a ninety-minute lecture.
• Built-in dictation. iPhone and Android both write down what the microphone hears live. It costs nothing and works in a quiet room, but it is designed for live speech rather than for a file you already recorded, and it stops when the screen sleeps.
• A transcription app. Fast, built for long files, and the free part differs enormously between products.

Before trusting a recording you cannot repeat to any of them, check two things: the maximum length allowed for a single file, which many free tools cap far below a full lecture, and whether the free tier asks for an account and a card up front. Then test on a short file whose content you can verify by ear.

Turning a raw transcript into notes you can revise from

A verbatim transcript is not notes. Two passes turn it into something usable.

Pass one: names and terms. Automatic transcription is weakest on proper nouns, acronyms and specialised vocabulary. Skim for the words that look wrong and correct them, and keep a small glossary of the course's recurring terms so each week goes faster. Do this first, because a wrong term propagates into the summary.

Pass two: structure. Add a heading whenever the topic changes, and mark whatever the lecturer repeated or called important.

Then produce the summary. In Zeepit you choose short for a reminder, medium for revision, or long when you missed the class and need the whole argument; a translation is there when the lecture was in another language. Copying the text or the summary is one tap and free, while exporting it as PDF, Word or plain text is part of Zeepit Pro and generated on the device.

What no pass fixes: several people talking over each other, and audio you cannot understand yourself.

What happens to your recording after you send it?

A lecture recording carries a named person's voice, and often other students' questions. It is worth knowing where the file travels.

Most services process audio on their servers, which is what makes a long file fast. Ask any tool the same three questions: is the file kept after processing, is it reused to train models, and where does the finished text live?

With Zeepit the file goes to Zeepit's servers, is transcribed, and is deleted right after. No archive, no reuse, no AI training on your recordings. The transcript and the summary stay on your device, which also means you should export or back up anything you want to keep.

Two habits apply whatever you use. Ask before recording someone: many universities and most workplaces require it, and asking costs nothing. And keep genuinely confidential material, medical, legal or HR, off any service you have not cleared with the person responsible for it.

Frequently asked questions