It depends on file size: Whisper takes 25 MB per file in the OpenAI API, while Potok accepts up to 2 GB. Whisper lists 98 languages, accuracy varies by language, and translates into English only; Potok covers about 30 languages, has no translation, and works as a Telegram bot.
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
- Whisper itself does not label speakers; a separate OpenAI model adds them, while Potok can switch labels on.
- Whisper's public pages do not name export formats; Potok exports TXT, MD, DOCX, PDF, SRT and VTT.
- Whisper's translation goes into English only; Potok has no translation of text.
- Whisper's public pages do not mention a Telegram bot; Potok lives in Telegram as @potok_now_bot.
- Speaker labels: Whisper lists Not in Whisper itself; a separate OpenAI model adds speaker labels, and Potok lists Yes, can be switched on.
At a glance
| Whisper | Potok | |
|---|---|---|
| Recognition languages | 98 languages, accuracy varies by language | About 30 |
| Translation of transcripts | Into English only | Not available |
| Speaker labels | Not in Whisper itself; a separate OpenAI model adds speaker labels | Yes, can be switched on |
| Export formats | Not stated on its public pages | TXT, MD, DOCX, PDF, SRT, VTT |
| Telegram bot | Not stated on its public pages | Yes, @potok_now_bot |
| Largest file | 25 MB per file in the OpenAI API | Up to 2 GB |
Facts about Whisper come from its public pages and from third-party reviews. They can change; check the vendor's site before you decide.
| 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 when choosing
| Mistake | What happens | Fix |
|---|---|---|
| Sending a long recording to the API unsplit | Files over 25 MB are refused by the OpenAI API. | Check the file against the per-file cap before uploading. |
| Expecting Whisper alone to name speakers | The text arrives as one block without who said what. | Use a speaker-label model or a tool that includes it. |
| Treating 98 languages as equal quality | Rare languages may produce noticeably weaker text. | Test your own language on a short clip first. |
| Expecting translation into languages beyond English | Whisper's translation stops at English, so other targets fail. | Decide the target language before choosing the tool. |
| Deciding from one line of the table alone | A strong line can hide a weak one that matters for your recording | Go through every row and mark the ones that are deal breakers for you |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
Why does accuracy vary between Whisper languages?
Whisper states 98 languages but says accuracy varies by language. Potok covers about 30 languages with good published results.
What does the 25 MB limit mean in practice?
A file above 25 MB must be split or compressed before it goes to the OpenAI API. Potok accepts one file up to 2 GB.
Can Whisper translate into any language?
No, its translation goes into English only. Potok does not translate text, so only the original language comes back.
Do I need code to try Potok?
No. Open @potok_now_bot in Telegram and send the file; the transcript returns in the chat with numbered speakers if labels are on.
Does Whisper tell speakers apart by itself?
Not in Whisper itself. A separate OpenAI model adds speaker labels, while Potok can switch them on directly.
What does the table say about largest file?
For Whisper it lists 25 MB per file in the OpenAI API, and for Potok it lists Up to 2 GB. Check the vendor page for the current wording before you decide.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Whisper, official page, Whisper · 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