- Which AI engine should I use?
- Fast (Lightning) is the best for English and other European languages — over 3,000x real-time on our GPU. Highest Accuracy, the default, makes the fewest errors in any of 99 languages; use it for legal, medical, or academic transcription where errors are costly. The Asian Languages engine is best for Chinese, Japanese, Korean, and other Asian languages.
- What languages are supported?
- 99 total: full coverage of European, Asian, Slavic, Romance, Semitic, and most South Asian / African languages. Auto-detect picks the right engine and language for you.
- What are word-level timestamps?
- Precise timing for each individual word, useful for karaoke-style subtitles or precise caption sync. Other modes give sentence-level timing.
- How fast is the auto-subtitle generator?
- Fast mode processes at over 3,000x real-time — a 10-minute video takes less than a second of compute. Highest Accuracy is slower but still typically finishes in seconds for short clips.
- How is this different from Submagic, Veed, or Kapwing?
- Three things. (1) Free with no watermark for SRT extraction — Submagic ($16/mo) and Veed limit free features behind their paywall. (2) Three AI engines — pick the fastest for English, the most accurate for any language, or the one built for Asian languages. (3) Burn subtitles directly into the video AND extract a portable SRT file — no extra step.
- Is this a free alternative to Submagic?
- Yes for the auto-caption step. Submagic's value-add is the curated animated caption presets and emoji auto-suggest — if you need those polished social-media-ready captions, check our Animated Captions tool which mimics Submagic's style for free.
