Upload the video file whole — we pull the audio track out on our side.
You convert a video to text by uploading the file directly — MP4, MOV, MKV, AVI, WMV and WebM all read the same way, we pull the audio track out of them, so you do not have to extract it first. The 5 GB limit is a real constraint here: an hour of 1080p is around 2.2 GB. We do not fetch videos from links, including YouTube.
Drop a file here or choose one from your device. We will show you the beginning of the transcript first.
skryba.ai transcribes on Scribe v2, which gets 2.2% of words wrong in Artificial Analysis' independent index — the lowest of every model measured. Gemini 3.5 Transcribe gets 2.6% wrong, Whisper large-v3 4.1%. It is the same model that writes the transcript preview you see before paying.
| Model | WER |
|---|---|
| skryba.ai (Scribe v2)the model we transcribe on | 2.2% |
| MAI-Transcribe-1.5Microsoft Azure | 2.4% |
| Gemini 3.5 TranscribeGoogle | 2.6% |
| Universal-3 ProAssemblyAI | 3.1% |
| GPT TranscribeOpenAI | 3.3% |
| Whisper large-v3OpenAI | 4.1% |
| Nova-3Deepgram | 5.2% |
Source: Artificial Analysis, AA-WER v2 index, non-streaming mode — read 27 August 2026. Seven of the 40-plus models measured: the top of the table and the names you already know. The index is measured mostly on English audio; for other languages, Polish included, no comparable public ranking exists. That is why we show you the start of every transcript before we ask for anything.
For audio the limit is theoretical — hundreds of hours fit. For video it is real, because the size is driven by the picture, which we do not use anyway.
| Material | Bitrate | Per hour | Fits in 5 GB |
|---|---|---|---|
| Online meeting recording, 720p | 1.5 Mbps | 0.7 GB | ~8 h |
| Typical 1080p | 5 Mbps | 2.2 GB | ~2.4 h |
| High-quality 1080p | 10 Mbps | 4.5 GB | ~1.2 h |
| Audio alone exported to MP3 | 128 kbps | 58 MB | ~93 h |
Scroll the table sideways to see the remaining columns.
The takeaway from the last row: for long, high-quality material, exporting just the audio track to MP3 shrinks the file by a factor of dozens and changes accuracy by nothing at all. The upload is faster too.
We do not support them, and we are not planning to work around that. Pasting an address from YouTube, Vimeo, TikTok, Facebook, SoundCloud or Spotify is rejected — those platforms require stream extraction, which we do not do.
What does work: pasting a direct HTTPS address to an audio or video file — from your own server, say, or a company drive that exposes a direct link. Those we fetch up to 500 MB.
A transcript with timestamps is what subtitles are made from. The SRT export loads into an editing suite or uploads straight to a video platform; the format is standard and needs no conversion.
No. Upload the file whole — we read the audio track out of it. Extracting only makes sense when the file exceeds 5 GB or you want a faster upload.
No. YouTube, Vimeo, TikTok, Facebook, SoundCloud and Spotify are rejected because they require stream extraction. Download the file and upload it — or paste a direct HTTPS address to a file, which we fetch up to 500 MB.
Any that carries an audio track — MP4, MOV, MKV, AVI, WMV, WebM, M4V, MPG, MPEG and 3GP among the common ones. MOV is what an iPhone records and what QuickTime exports, so upload it as it is, no conversion. If the file is over 5 GB, export just the audio track to MP3, which shrinks it anyway.
Yes, SRT export is available on paid plans. You see the timestamped transcript before that — in the free preview and the 3-day trial — so you can judge the quality before buying.
