Russian speech to text in Potok is recommended, and it works differently from other languages: a model built for Russian speech handles it. Published figures for the general model therefore do not apply, so we quote none. Send voice, audio or video and export in one of 6 formats.
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
- We cite no error rate for Russian, because the figures others published describe the general model, not this one.
- Without a number, the sensible test is your own: send a short clip from the real speaker and read it.
- Russian text is written in the Cyrillic script, and the exported files keep it as spoken.
- Stress is not marked in normal text, so words that differ only by stress are chosen by context.
What a transcript looks like
[00:00] Speaker 1: Давайте начнём встречу. Сначала разберём итоги прошлой недели, потом спланируем следующие шаги.Meaning in English: Let's start the meeting. First, we'll review last week's results, then plan the next steps.
Written by us to show the script and the layout of a transcript. It is not the output of the recognition engine.
Published quality
| Language | Verdict | Published result | Notes |
|---|---|---|---|
| Russian (Русский) | Recommended | No published figure | A model built for Russian speech is used for this language, so the published figures for other models do not apply. |
These numbers come from published tests by other people, not from tests by Potok. Your own audio, with its noise, accents and microphone, will differ.
How to do it
- Run a short test. Send a minute of the real speaker first. With no published score, your own clip is the best evidence.
- Message the bot. Open @potok_now_bot in Telegram and send the file, a forwarded voice message, or a YouTube link.
- Bring the long version. Up to 2 GB can go in, and the Upload button in the mini app takes the biggest.
- Check foreign words. Brand names and borrowed terms are where a Russian-specific model may write something unexpected.
- Review case and stress. Read endings and any word whose meaning depends on stress.
- Pick an export. TXT, MD, DOCX, PDF, SRT or VTT make six formats; short results also appear in the chat.
| 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 |
|---|---|---|
| Looking for an error rate to cite | None applies to this model. | Describe your own test instead. |
| Applying numbers from another language's page | They come from a different model. | Leave them out of your notes. |
| Ignoring foreign names in the text | Spelling of borrowed words varies. | Search for them and fix by hand. |
| Skipping timestamps for video work | The text will not line up with the picture. | Turn timestamps on and export SRT or VTT. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
What is the error rate for Russian?
We give none, because published figures for the general model do not apply to the Russian-specific one.
Why is Russian different?
It is recognised by a model built specifically for Russian speech, not by the model behind the other languages.
How do I judge the quality?
Send a short clip from your own speaker and read the transcript against the audio.
Does the bot change the language of the text?
No. The transcript stays in Russian, in the same language as your recording, and nothing else is produced.
How many export formats are there?
Six: TXT, MD, DOCX, PDF, SRT and VTT.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Word error rate (Wikipedia), Wikipedia · checked
- Whisper speech recognition system (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