What Is Real-Time Transcription? Live Text Explained

What is real-time transcription? It is transcription that appears while a person is still speaking, a few words behind the voice. The text is produced in small pieces from a live audio stream, and earlier words may be corrected as more sound arrives. Live captions in meetings work this way.

Transcribe a recording

Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.

Open @potok_now_bot in Telegram

Files up to 2 GB · export to TXT, MD, DOCX, PDF, SRT and VTT · speaker labels and timestamps · about 30 languages

Key facts

The system must answer quickly, so it sees only the words spoken so far and cannot use what comes next to settle an unclear word.
Partial results change on screen: a word shown a moment ago may be replaced once the sentence is finished.
Live use has no second pass, so punctuation, names and numbers are usually weaker than in a transcript made from a finished file.
The opposite is batch transcription, where a complete recording is processed after the event and the whole context is available.
A stable connection and a close microphone matter more here, because a dropped second of audio cannot be replayed.

What it looks like

// a stream of partial results as one sentence is spoken
partial  0.8 s  "the meeting"
partial  1.6 s  "the meeting starts at"
partial  2.4 s  "the meeting starts at nine"
final    3.1 s  "The meeting starts at nine."

See the full list of languages, the tools and the glossary.

Frequently asked questions

What is the difference between real-time and batch transcription?

Real-time shows text during the speech from a stream. Batch processes a finished file afterwards, which allows more context and a more careful result.

Why does live text keep changing?

The system revises early guesses when later words make the meaning clear. This is normal for partial results and settles once the sentence ends.

Is live transcription good enough to publish?

Treat it as a draft. For anything you publish, re-run the saved recording as a file and proofread the output.

How do I transcribe a recording I already have with Potok?

Open the bot, send the file as audio or as a document, or forward a voice message. Files up to 2 GB are accepted.

What improves live results most?

A good microphone close to the speaker, little background noise and one person talking at a time.

Sources

Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.

Speech recognition (Wikipedia), Wikipedia · checked 2026-10-18
Closed captioning (Wikipedia), Wikipedia · checked 2026-10-18

Try it on your own recording