AI Transcription

Transcribe podcasts, interviews, lectures, and meetings to clean text. AI-powered, audio + video inputs, exports to plain text, SRT, VTT, and DOCX.

or press Ctrl+V to paste

Free to transcribe — pay only to download the final transcript (TXT, SRT, VTT, or DOCX).

How to Use AI Transcription

  1. Upload a podcast, interview, lecture, or meeting recording
  2. Choose the source language (or leave on auto-detect)
  3. Click Transcribe and wait while our AI runs on a GPU
  4. Read the transcript inline, then download as TXT / SRT / VTT / DOCX

Features

  • Transcribe audio and video in 99+ languages
  • Exports: plain text, SRT, VTT, and Word DOCX
  • Automatic punctuation, casing, and filler-word removal
  • Designed for podcasters, journalists, lawyers, and researchers
  • GPU-accelerated — 10-minute clip transcribed in under a minute
  • Free to use — pay only to download the final transcript

Frequently Asked Questions

What file formats can I upload?
Any audio (MP3, WAV, M4A, FLAC, OGG, OPUS) or video (MP4, MOV, MKV, WebM, AVI). We extract the audio track automatically.
How accurate is the transcription?
Around 95-99% on clear audio — comparable to commercial services like Otter, Rev, and Trint. Accuracy drops on heavy background noise, strong accents, or overlapping speakers.
What's the difference between AI Transcription and AI Subtitles?
AI Subtitles is optimized for video subtitles — you get an SRT file or a video with burned-in captions. AI Transcription is optimized for the transcript itself — you get a clean, paragraph-formatted document (plain text, Word DOCX, SRT, or VTT) designed for reading, quoting, and editing.
How do I export the transcript to Microsoft Word?
After transcription completes, click Download as DOCX. The file opens directly in Microsoft Word, Google Docs, or any word processor that supports the .docx format.
Are speaker labels supported?
Speaker diarization (auto-labelling who said what) is coming soon as a Pro/Lite feature. For now, the transcript is a single-speaker stream — you can manually add speaker labels in your editor after download.
Is my audio kept private?
Yes. Files are processed on our own servers (not shared with third parties), uploaded transcripts auto-delete after 24-72 hours depending on plan, and we never use your audio for training.