Paste a URL (YouTube, podcast, Dropbox) or upload a file. We transcribe in minutes with speaker labels and word-level timestamps. First 30 minutes free.
Three steps. From paste to transcript in minutes.
YouTube / podcast URL, Dropbox / Google Drive link, or a file from your device.
Server-side Whisper-class models with speaker diarization. Most 1-hour recordings complete in 2–5 minutes.
Word-click to seek, edit errors, chat with the transcript, export as TXT / SRT / VTT / DOCX.
First 30 minutes of audio are free. After that it's $1 per hour of audio, pay-as-you-go. No subscription, no card needed until you exceed the free tier.
YouTube videos, podcast episodes, Dropbox / Google Drive / S3 links, and direct uploads (MP3, MP4, WAV, M4A, MOV, MKV, and more — anything ffmpeg ingests).
Yes — speaker diarization runs on every job by default, free tier included. Most competitors either charge for it or don't support it on free / starter tiers.
100+ languages via Whisper-class models. The top-20 by usage (English, Spanish, French, German, Portuguese, Italian, Dutch, Hindi, Mandarin, Japanese, Korean, Arabic, Russian, Turkish, Polish, Ukrainian, Vietnamese, Indonesian, Thai, Hebrew) are all high-quality.
Yes — export as SRT, VTT, DOCX, or plain TXT from the transcript view. Or use our dedicated subtitle tool.
Paste a URL or drop a file. First 30 minutes free, no card needed.
Start transcribing →