To transcribe a news interview, send the recording to the Potok bot on Telegram with speaker labels and timestamps switched on. Files up to 2 GB work. You get a draft you can search by name, then check each line you plan to quote against the sound.
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
- Exact quotes need a second listen: the draft points you to the spot, and your ears confirm the wording.
- A timestamp on each passage lets you jump to a quote in the recording without scrubbing through the whole file.
- Labels such as Speaker 1 mark who talks, but only you know whether that is the minister or the aide.
How to do it
- Note who is speaking. Before you record, write down each voice in order of appearance. You will need that list to turn Speaker 1 into a name later.
- Record with a backup. Run a second recorder close to the guest. If the main file has a dropout, the backup saves the quote.
- Open the bot. Open @potok_now_bot in Telegram and start a chat with it.
- Switch on speaker labels and timestamps. Do it before you send the audio, so every passage carries a time and a speaker number.
- Send the recording. Attach the audio or video as a file. A long press conference can reach 2 GB; use the Upload button of the mini app for that size.
- Mark the quotes you need. Read the draft once and highlight lines you may use. Copy the time next to each so you can find it again.
- Verify each quote by ear. Play the exact seconds around every highlighted line. Fix the wording, and keep the speaker's own grammar if you quote them.
- Export for the desk. Pick DOCX if an editor will comment on it, or TXT for your notes. Keep the timestamps in the file you hand over.
What the result looks like
[00:00] Speaker 1: Thank you for joining us. Why did the council delay the vote?
[00:06] Speaker 2: We needed the traffic report first, and it arrived on Monday.
[00:12] Speaker 1: So the road plan is not final?
[00:15] Speaker 2: Correct. Nothing is final until the council votes.
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 |
|---|---|---|
| Printing a quote straight from the draft | One wrong word can change the meaning and put words in the guest's mouth. | Listen to the original seconds for every quote before publication. |
| Trusting a speaker label as a name | You could credit a statement to the wrong official. | Match each number to a person yourself, using your notes from the interview. |
| Cleaning up spoken grammar inside quotation marks | The text no longer matches the recording and can be challenged. | Quote as spoken, or paraphrase outside quotation marks. |
| Ignoring numbers and titles | Figures, ranks and place names are the usual weak spots in a draft. | Check every figure and job title against the audio and a second source. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
Can I use the transcript as a source for direct quotes?
Use it as a finder, not as the final word. Confirm the exact wording by listening to the recording before you print it.
Does the bot know who the speakers are?
No. It numbers the voices as Speaker 1, Speaker 2 and so on. You add the real names from your own notes.
How do I find a quote I remember hearing?
Switch on timestamps and search the text for a key word. The time next to the line shows where to listen.
Is a video interview supported?
Yes. Send the video as a file and the bot reads the sound track. You can export SRT or VTT if you want captions.
What should I do about a noisy street interview?
Expect more mistakes and read the draft with the audio playing. Get the microphone near the mouth next time.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Speaker diarisation (Wikipedia), Wikipedia · checked
- Transcription (linguistics) (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