Free Japanese transcription should be tried on a clip before it is trusted. Japanese has no spaces, so quality is quoted as a character error rate: published tests of others give 4.5 for large-v3, which Potok uses. Read a short sample against the audio and check the kanji.
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
- Every Japanese score quoted here counts characters, and it was produced by other people on their own test recordings.
- Without spaces, a word error rate has nothing to count, which is why characters are the unit here.
- Mixed writing means the same sound can be spelled with kanji, hiragana or katakana, so some differences are choices, not errors.
- A tool can hear a word right and still pick a different kanji for it.
What a transcript looks like
[00:00] Speaker 1: 会議を始めましょう。まず先週の結果を確認し、そのあと次のステップを計画します。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 |
|---|---|---|---|
| Japanese (日本語) | Recommended | character error rate 4.5% (large-v3, FLEURS) Whale paper | Published tests show a low error rate for this language. Japanese is written without spaces, so quality is reported as a character error rate, not a word error rate. |
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
- Choose a clip with homophones. Pick speech that has names or technical terms, where several kanji share one reading.
- Open the bot. Open @potok_now_bot in Telegram and send the clip as a voice message, an audio file or a video.
- Read character by character. Do not count words. Mark each wrong, missing or extra character and divide by the length of the text.
- Separate sound from spelling. If a word reads correctly but uses another kanji, note it as a spelling issue and decide whether it matters for you.
- Export for review. Save TXT or DOCX to proofread, or SRT if you need subtitles.
| 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 |
|---|---|---|
| Counting words in Japanese | There are no spaces to count, so the tally comes out meaningless. | Count characters, or sentences with a mistake. |
| Treating a kanji choice as a hearing failure | You may blame the tool for something a human editor settles quickly. | Mark spelling and sound errors separately. |
| Testing only polite, scripted speech | Casual talk with dropped particles behaves differently. | Add a clip of relaxed conversation. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
Why is a character error rate used?
Japanese is written without spaces, so characters are the natural unit to count.
Is the 4.5 your own result?
No. It comes from published tests of others, not from our measurements.
Will furigana or romaji appear?
Do not expect them. The text comes back as ordinary Japanese writing.
Can I get an English version?
No. The transcript stays in the language of the recording.
Can I send a long recording?
Yes, files up to 2 GB are accepted, using the Upload button of the mini app for large ones.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Japanese writing system (Wikipedia), Wikipedia · checked
- Word error rate (Wikipedia), Wikipedia · checked
- Whale paper, Table 5 (Whisper large-v3 on FLEURS) · 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