To convert French audio to text, send it to the Potok bot on Telegram. Published tests of others on large-v3, not our measurements, give a word error rate of 5.3, with reported results from 5.3 to 6.31. Files up to 2 GB are accepted, and the text stays in French.
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
- French has many words that sound alike but are spelled differently, so context decides the spelling.
- Silent endings for plural and feminine forms are the commonest small slips.
- Accents on vowels change meaning, so check them in names.
- Switch on speaker labels for interviews and debates.
- Quoted passages in other languages inside a French talk are the least reliable lines.
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
- Open the bot. Open @potok_now_bot in Telegram.
- Send the file. Audio, video, voice note or YouTube link, up to 2 GB. Use Upload for big files.
- Scan sound-alike words. Words that sound the same need a human check against the context of the sentence.
- Check accents and endings. A missing accent or silent ending can change a word.
- Listen again to numbers and dates. French numbers are built in unusual ways, so check amounts and dates against the recording before you pass the text on.
- Export. DOCX, TXT, PDF, MD, 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 |
|---|---|---|
| Trusting sound-alike words | They are chosen from context and sometimes wrongly. | Read for meaning, not only for sound. |
| Ignoring agreement endings | Silent endings are easy to miss. | Check plurals and feminine forms. |
| Sending low-quality phone clips | Compression removes detail. | Send originals. |
| Mixing in English terms | Loan words may be spelled French style. | Fix them by hand. |
| Judging liaison by ear only | Linked sounds between words can hide a word boundary, so two words may be run together. | Re-listen to any phrase that reads oddly. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
How good is French?
The published word error rate is 5.3 on large-v3, among the best on our list.
Why a range?
Different tests report values from 5.3 to 6.31 for the same model.
Does it handle Canadian French?
The published numbers do not separate regional varieties, so test a sample.
Is there a size limit?
2 GB per file, with Upload for the largest.
Is the text translated?
No. The transcript stays in French, the language of the recording you sent to the bot.
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
- FLEURS speech dataset (Google, Hugging Face), Hugging Face · 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