Is the output in Cyrillic?

Yes, standard Russian orthography with punctuation — not transliteration.

Does it work with Ukrainian or Belarusian audio?

Those are separate languages in the model and should be selected explicitly; Ukrainian is available in the language list.

Is there a file-size limit?

Files up to 8 GB are accepted. The video is read from disk, so what uses memory is the extracted audio, about 230 MB per hour of recording. Nothing is uploaded in the default Private mode.

How are English words in a Russian talk written?

Inconsistently. A term such as deadline or feature may appear in Latin letters in one sentence and in Cyrillic in the next, which is normal for the model. A quick find-and-replace afterwards standardises them.

Does it handle Russian from Kazakhstan, Ukraine or the Baltics?

Yes. Regional pronunciation of Russian is handled well, since the language itself is the same. The difficulty arises only when speakers switch languages mid-sentence, where words from the other language may be misread.

Transcribe Russian Audio to Text — Free, No Upload

Convert Russian audio or video into Cyrillic text in your browser. 100% free, no per-minute limit, no account, processed on your device.

Model

Spoken

Output

Timestamps [0:05]

⬇️ One-time download: 104 MB (Accurate model 73 MB + 31 MB audio engine, shared with the video tools) — only the first time: it stays saved in your browser, so next time it starts right away. (Translating or switching model needs its own file; a browser can clear this cache if storage runs very low.) Processing takes about 4-8 min per 10 minutes of audio on a typical laptop.

What exactly is downloaded — and is it safe?

• The speech model is OpenAI's Whisper, open source, converted to run in browsers. The audio engine is FFmpeg, the open-source tool behind most video software.
• They come from Hugging Face and unpkg — public repositories used by researchers and developers worldwide, not from us.
• They are not programs. Nothing is installed, no permission is asked, and your browser keeps them in its own cache exactly like it keeps a website's images.
• They can't read your files or touch anything outside this page. Remove them any time by clearing this site's data in your browser.
• Your audio never goes the other way: this site has no server receiving files.

Drop your video or audio here or click to choose

transcribed on your device by OpenAI's Whisper — your recording is not uploaded

ℹ️ Use a recording on your device that you own or have permission to process. This tool cannot fetch YouTube or social video links. Start with a short clip. Larger models can improve accuracy but require more memory and download data. Processing time varies by device, model and recording length. Always review the result.

Russian audio arrives from a wide spread of sources: online lectures and courses, interviews and podcasts, dictated notes, voice messages in Telegram and WhatsApp, recorded consultations and the lessons of people studying the language. This page fixes the language to Russian before you start, so a short or noisy clip is not misread as Ukrainian or Bulgarian by the detector, and the result is Cyrillic text with punctuation rather than a transliteration.

What the model does well is standard spoken Russian at a normal pace, including speakers from outside Russia. Where it needs review: proper names and surnames, which it spells by ear; figures and dates in a financial or legal recording; and technical talks that mix in English terms, which can come out in either alphabet from one sentence to the next. As in most printed Russian, the letter ё is generally written as е. If you need the content in English rather than in the original, there is a translation option, and timed subtitle files can be exported for video. Pick the Fast model to get through a long lecture quickly and the Best model when accuracy matters more than time. In the default Private mode the recording is processed in your browser and never uploaded, which matters when the audio is personal.

More sizes & options

Frequently asked questions

Is the output in Cyrillic?: Yes, standard Russian orthography with punctuation — not transliteration.
Does it work with Ukrainian or Belarusian audio?: Those are separate languages in the model and should be selected explicitly; Ukrainian is available in the language list.
Is there a file-size limit?: Files up to 8 GB are accepted. The video is read from disk, so what uses memory is the extracted audio, about 230 MB per hour of recording. Nothing is uploaded in the default Private mode.
How are English words in a Russian talk written?: Inconsistently. A term such as deadline or feature may appear in Latin letters in one sentence and in Cyrillic in the next, which is normal for the model. A quick find-and-replace afterwards standardises them.
Does it handle Russian from Kazakhstan, Ukraine or the Baltics?: Yes. Regional pronunciation of Russian is handled well, since the language itself is the same. The difficulty arises only when speakers switch languages mid-sentence, where words from the other language may be misread.

Guides

Related tools

🔒 Private mode (the default): your recording never leaves your device. Instant mode is optional, you switch it on yourself, and it sends the audio to the browser's speech service.