How audio and video transcription works in Hushscript

How it works

1. 01

   Preview one recording

   Before signing in, choose one local audio or video file, paste a supported audio link, or record in the browser. Hushscript makes a free speaker-labeled preview of up to 5 minutes with automatic language detection. For a local file or recording, only the preview clip is sent; an audio link is fetched and clipped by the server. Video links are not supported. After you sign in, a new personal account's first uploaded file gets its own free preview of up to 5 minutes with the recognition options you chose.

2. 02

   Sign in and add the full recording

   In the app, add a local audio or video file up to 10 hours long, record, or import a supported audio URL. Compatible audio uploads unchanged; audio from video is prepared on your device, so the source video stays there. Audio links are fetched by the server and use personal minutes, even in Teams. Stage local files as a batch or merge related ones in your chosen order; start the batch when ready.

3. 03

   Set transcription options

   Choose a source language or auto-detect, speaker detection, and multichannel audio. Use saved dictionaries and prompt presets with one-off keyterms or a prompt. Turn on medical mode or translation if you need them: the exact minute cost always shows before you confirm (see Pricing for the full breakdown). Private mode accepts local files or recordings, not URL imports, and has no medical mode or translation.

4. 04

   Transcribe and edit

   Hushscript transcribes prepared audio online and adds speaker labels when you select speaker detection. In normal mode, rename speakers, reassign or split and merge transcript turns, correct text, add tags, or find and replace. The editable transcript is saved for later review.

5. 05

   Generate optional Insights

   The first Insights generation per transcript is free in normal mode. One free run creates summaries, actions, decisions, open questions, chapters and a fully rewritten, cleaned-up transcript. Run it after transcription or choose automatic generation. Regenerating later costs minutes; Pricing lists the rates. Private mode skips Insights.

6. 06

   Export the result

   In normal mode, preview, copy or download 21 export formats without a watermark, including SRT and VTT subtitles and FCPXML and EDL timelines. Export packs bundle several formats; a password-protected .husharchive is separate from the 21. In private mode, you create a password-protected archive from a short-lived result without saving a transcript to your account.

7. 07

   Choose retention and deletion

   For personal accounts in normal mode, set a default to keep saved transcripts until you delete them or auto-delete after 7, 30, 90 or 365 days; you can override one transcript. Hushscript removes transcription audio after processing. Dedicated transcript, AI-artifact and account deletion flows return downloadable receipts. Private mode saves no account transcript; its short-lived result expires if you do not download and acknowledge it.