Free Bulgarian transcription: check a tool on a short file

By the Potok teamUpdated ~2 min read

To check a free Bulgarian transcription tool, send it a short clip with names and numbers in it and compare the text with what you hear. For Potok, published tests of the large-v3 model report a word error rate of 12.86 for Bulgarian, measured by others, not by us.

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 12.86 is a word error rate for large-v3 from a published test, not our own measurement.
  • Bulgarian is written in the Cyrillic alphabet, so a Latin-letter result means something went wrong with the language choice.
  • Names, places and numbers are where recognisers slip first, so put them in the test clip.
  • The transcript stays in Bulgarian and is not translated.

What a transcript looks like

Sample sentence in Bulgarian
[00:00] Speaker 1: Да започнем срещата. Първо ще прегледаме резултатите от миналата седмица, а след това ще планираме следващите стъпки.

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

Quality for this language
LanguageVerdictPublished resultNotes
Bulgarian (Български)Recommendedword error rate 12.86% (large-v3, FLEURS)
Canary-1B-v2 paper
Published tests show a low error rate for this language.

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

  1. Pick a clip with hard items. Choose a minute of speech that includes a few personal names, a place and a number or two.
  2. Open the bot. Open @potok_now_bot in Telegram and send the clip as a voice message or an audio file.
  3. Check the script first. The text should arrive in Cyrillic letters. If it does not, the wrong language was recognised.
  4. Mark the slips. Underline every name or number that differs from the audio. Count them against the words you read.
  5. Try a longer file. Files up to 2 GB work, and the Upload button of the bot's mini app is there for large ones. Export TXT or DOCX when you are happy.
What Potok can do
FeatureDetails
Where it worksTelegram bot @potok_now_bot; large files go through the bot's mini app
File sizeUp to 2 GB
ExportTXT, MD, DOCX, PDF, SRT, VTT
Speakers and timeSpeaker labels and timestamps can be switched on
LanguagesAbout 30 languages with good published results; a few more with warnings
TranslationThe transcript is not translated into another language
DictationA separate Potok desktop app lets you dictate text into any program; it recognises Russian speech

Common mistakes

Common mistakes and how to fix them
MistakeWhat happensFix
Testing with a clip full of silence and musicPauses and background music hide how the tool handles speech.Cut the clip so that people talk most of the time.
Ignoring numbers in the resultA wrong figure in a quote or an address can pass unnoticed.Read every number against the audio, one by one.
Treating 12.86 as the score for your fileThe figure comes from a published test on other recordings.Use your own clip to see how your voices behave.
Sending a poor phone recording for the first tryNoise masks differences between tools.Record a clean sample in a quiet room, then test noisy audio second.

More

See the full list of languages, the tools and the glossary.

Frequently asked questions

Is Bulgarian among the recommended languages?

Yes, the published results for Bulgarian are good, with a word error rate of 12.86 for large-v3 in the test we rely on.

What does a word error rate of 12.86 mean?

Roughly that about thirteen words in a hundred differ from a human reference in that test. Your audio can score better or worse.

Can I check the text against the audio easily?

Yes. Turn timestamps on in the settings, then jump to the passage in the recording to confirm a doubtful word.

Does the bot accept video?

Yes. Audio, voice messages, video files and YouTube links are accepted within the 2 GB limit.

Will I get an English version?

No. The transcript stays in the language of the recording.

Sources

Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.

  1. Word error rate (Wikipedia), Wikipedia · checked
  2. Whisper speech recognition system (Wikipedia), Wikipedia · checked
  3. Canary-1B-v2 paper, Table 11 (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