- 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.