How accurate is the AI transcription?

Our AI achieves over 95% accuracy for clear audio in most languages. Background noise may reduce accuracy.

Which languages are supported?

Over 100 languages including English, French, Spanish, German, Arabic, Chinese, Japanese, and more.

What output formats are available?

You get SRT (SubRip), VTT (WebVTT), and plain TXT files bundled in a ZIP download.

Is there a file size limit?

Up to 500 MB with a free account, 2 GB with Pro. Audio and video files are accepted.

Can I use the transcription for subtitles?

Yes! The SRT and VTT files can be imported into any video editor or uploaded to YouTube, TikTok, etc.

How long does transcription take?

Typically 1-3 minutes for a 10-minute video. Longer videos take proportionally more time.

Can I edit the transcript before exporting?

The exported files can be opened in any text editor for corrections. For on-platform editing before burning subtitles, Pro users can review the transcript in the subtitle burn tool.

Does it handle multiple speakers?

The current version does not separate speakers (diarization). It produces a single continuous transcript. Speaker identification is on our roadmap for a future update.

Can I transcribe a live stream or URL?

You need to upload a file directly. For TikTok or Instagram videos, download the video first, then upload it here for transcription.

Does transcription work well with regional accents?

Yes, the Whisper model handles regional accents very well across most languages. For best results, remove background noise before transcription to improve voice clarity.