Potok transcribes Indonesian voice messages, audio and video through a Telegram bot. The published word error rate is 7.1, a good result, though it comes from the earlier large-v2 model because no large-v3 figure was found. Send recordings and files up to 2 GB.
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
- A word error rate of 7.1 comes from published tests of the earlier large-v2 model, not from our own runs.
- A third party published a turbo average of 6.98, which is close to the older figure.
- No large-v3 result for Indonesian was found, so the figure may not match the current model.
- Casual spoken Indonesian differs from formal written style, and the text may mix them.
What a transcript looks like
[00:00] Speaker 1: Mari kita mulai rapat. Pertama, kita tinjau hasil minggu lalu, lalu kita rencanakan langkah berikutnya.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 |
|---|---|---|---|
| Indonesian (Bahasa Indonesia) | Recommended | word error rate 7.1% (large-v2, FLEURS) Whisper paper, TheWhisper benchmark | Published tests show a low error rate for this language. No large-v3 figure was found; the figure is from the earlier large-v2 model, and a third party published a turbo average of 6.98. |
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
- Go to the bot. Open @potok_now_bot and send the Indonesian file, voice note or link.
- Choose the upload route. Up to 2 GB can go in the chat; for a long meeting use the Upload button of the mini app.
- Review informal words. Slang and shortened words may be written in an unexpected form.
- Review repeated words. Plurals are often made by writing a word twice with a hyphen, so look for a missing one.
- Export. TXT, MD, DOCX, PDF, SRT or VTT.
| 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 |
|---|---|---|
| Assuming the figure comes from the latest model | The result for large-v3 may differ. | State that the figure is for large-v2. |
| Missing hyphenated plurals | Words such as buku-buku appear with one half missing. | Search for single words that should be doubled. |
| Mixing regional languages into the recording | Words from other regional languages are not covered by the figure. | Check those lines by hand. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
How good is Indonesian transcription?
Published tests of the earlier large-v2 model give a word error rate of 7.1 for Indonesian, which is a good result on our list.
Why does the note mention large-v2?
We found no large-v3 figure for Indonesian, so the available number belongs to the older model.
Does it work for casual chat recordings?
Yes, though slang and shortened words are more likely to need corrections.
What if the speaker uses a regional language too?
The published figure does not cover that mixed speech, so review those passages carefully.
Is the text translated?
No. The text stays in Indonesian, and the bot does not offer a translation into another language.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- FLEURS speech dataset (Google, Hugging Face), Hugging Face · checked
- Whisper speech recognition system (Wikipedia), Wikipedia · checked
- Whisper paper, Table 13 (large-v2 on FLEURS) · checked
- TheWhisper benchmark (turbo, average over FLEURS, MLS, Common Voice) · 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