To transcribe a radio show, send the broadcast recording to the Potok bot on Telegram with speaker labels and timestamps on. Files up to 2 GB work, enough for a long programme. Expect gaps where music plays, and read those parts with the audio playing.
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
- Music sections are the hard part: songs and jingles can produce stray words or leave blank stretches.
- Several speakers means a presenter, callers and guests, so the numbered labels need a careful clean-up.
- A timestamp at each segment helps archivists catalogue the show by topic.
How to do it
- Pick the right recording. Use the broadcast file rather than a recording made through speakers. A clean line feed gives better words.
- List the voices. Write down the presenter, regular guests and any callers, so labels can be matched later.
- Open the bot. Open @potok_now_bot in Telegram and start a chat with it.
- Turn on labels and timestamps. Both help when a two-hour show has many segments and changes of voice.
- Send the file. Attach the broadcast as a file. For a very large recording, up to 2 GB, use the Upload button of the mini app.
- Mark the music sections. Replace stray words in songs with a note such as [music]. Keep the time so the segment can be found.
- Catalogue the text. Export DOCX or TXT and add the date, programme name and topics in your own cataloguing system.
What the result looks like
[00:00] Speaker 1: Good morning, you are listening to the local news hour.
[00:05] Speaker 1: First, the weather, then a song from our request list.
[00:11] Speaker 2: [music]
[03:40] Speaker 1: Welcome back. Our guest today runs the town library.
Sample text written by us to show the layout of a transcript. It is not the output of the recognition engine and not a real recording.
| Feature | Details |
|---|---|
| Where it works | Telegram bot @potok_now_bot; large files go through the bot's mini app |
| File size | Up to 2 GB |
| Export | TXT, MD, DOCX, PDF, SRT, VTT |
| Speakers and time | Speaker labels and timestamps can be switched on |
| Languages | About 30 languages with good published results; a few more with warnings |
| Translation | The transcript is not translated into another language |
| Dictation | A separate Potok desktop app lets you dictate text into any program; it recognises Russian speech |
Common mistakes
| Mistake | What happens | Fix |
|---|---|---|
| Keeping lyrics that are really noise | Random words from a song pollute the archive text. | Replace music passages with a short bracketed note. |
| Mixing up callers | A caller's claim ends up credited to the presenter. | Check labels at each change of voice, especially after a phone line. |
| Recording from a radio speaker | Room echo and noise lower the quality of the words. | Use the direct broadcast file whenever you can. |
| Forgetting rights | Publishing quoted songs or interviews may need permission. | Check with the station or the rights holder before you post the text. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
What happens to the music in the recording?
Songs may give stray words or empty stretches. Check these sections and mark them with a note in the final text.
Can I transcribe a full-day broadcast?
Cut it into shows or hours. Each file can be up to 2 GB, and smaller files are easier to review.
Does it separate callers from the presenter?
It numbers the voices it hears. Callers on a phone line sound different, so check those changes by ear.
Which format is useful for an archive?
TXT is a safe plain copy, and DOCX is good while you are editing. Keep timestamps so entries can be searched by segment.
Are broadcast recordings protected?
Many are. Check the rules and the station's permission before you publish text from a programme.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Voice activity detection (Wikipedia), Wikipedia · checked
- Signal-to-noise ratio (Wikipedia), Wikipedia · checked
Try it on your own 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