Potok transcribes French voice messages, audio and video in a Telegram chat. Published tests of the large-v3 model report word error rates from 5.3 to 6.31 for French, a low range among the languages we list. Files up to 2 GB are accepted.
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
- The published range for French is 5.3 to 6.31 word error rate, measured by others on large-v3.
- Three sources agree on a low result, which is why French has no warning.
- Many French words sound the same but are written differently, so grammar endings are the typical remaining mistakes.
- The text is returned in French and is not translated.
What a transcript looks like
[00:00] Speaker 1: Commençons la réunion. D'abord, nous passerons en revue les résultats de la semaine dernière, puis nous planifierons les prochaines étapes.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 |
|---|---|---|---|
| French (Français) | Recommended | word error rate 5.3% (large-v3, FLEURS) Canary-1B-v2 paper, Whale paper, TheWhisper benchmark | Published tests show a low error rate for this language. Published large-v3 results range from 5.3 to 6.31. |
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
- Hand over the audio. Send your French recording to @potok_now_bot, as a voice message, file, video or YouTube link.
- Big file route. Up to 2 GB can be attached; the Upload button of the mini app suits very long recordings.
- Choose labels. For an interview, enable speaker labels and timestamps so each turn is marked.
- Read for agreement. Check endings such as plural and past participle forms, where the sound gives no clue.
- Check accents and apostrophes. Look at é, è, ê, ç and elisions like l'homme.
- Export. TXT, MD, DOCX and PDF for documents; SRT and VTT for French 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 |
|---|---|---|
| Trusting silent endings | A plural s or a feminine e may be missing, because they are not heard. | Read for grammar, not for sound. |
| Assuming accents from France and Canada behave the same | The published range does not break them down. | Test a short clip from your speakers. |
| Skipping a check on homophones | Words such as ou and où can be swapped. | Search for the pairs that matter to you. |
| Looking for an English version | The bot does not translate the transcript. | Translate separately. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
How good is French transcription?
Published tests of the large-v3 model give a word error rate of 5.3 to 6.31 for French. Other groups measured these numbers, not us.
Why is it a range?
Different test sets and text rules give slightly different results, so we show the published range.
Are French accent marks kept?
Yes in general, but review them. A wrong accent can change the word.
Can I send a French podcast episode?
Yes. Send the audio file within the 2 GB limit, or use the Upload button for a very large one.
Are speakers identified by name?
No. They are labelled Speaker 1, Speaker 2 and so on.
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
- Canary-1B-v2 paper, Table 11 (Whisper large-v3 on FLEURS) · checked
- Whale paper, Table 5 (Whisper large-v3 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