Convert video to text from a local file

Hushscript is a video transcriber and video to text converter for local files. It prepares the video’s audio track in your browser, so only the prepared audio reaches the transcription service and the source video stays on your device. The result is an editable, timecoded video transcript you can search, correct, and export for documents, subtitles, translations, or editor timelines.

Your video never leaves your device

Choose a local MP4, MOV, WebM, or MKV file and preparation starts in your browser. A compatible audio track is copied without re-encoding; files that need conversion take the local conversion path. The source video stays on your device throughout, only prepared audio is sent for transcription, and Hushscript removes its transcription-audio copy after processing, never trains on your recordings, and encrypts stored transcript content at rest.

Transcribe video to text in three steps

To turn video into a transcript, choose the local video and preview up to 5 minutes before creating an account. The preview uses automatic language detection. In the full app, choose automatic detection or set the source language, adjust speaker and vocabulary controls, and merge related recordings into one ordered transcript when a session spans several files. Speaker detection separates voices where the language supports it, and you can rename or reassign the labels while editing. Machine transcripts can contain errors, so check names and numbers against the recording before you rely on them.

Turn the transcript into subtitles

Once the video is transcribed, Hushscript’s subtitle generator can export SRT, VTT, ASS, TTML, SBV, or LRC from the timed transcript. These are subtitle files you can use in another player or editor; Hushscript does not burn captions into the source video or edit the footage on this page. FCPXML and EDL exports carry transcript timing into compatible editing workflows. For a step-by-step, see how to create SRT subtitles from a video.

Translate the transcript into other languages

Choose one or more of 13 target languages before transcription, and translation runs in the same job. The source can be one of roughly 99 transcription languages; each selected target is generated from the original transcript and adds 25% of the recording duration to the billed minutes. Translated transcripts can use the same 21 export formats, including SRT and VTT subtitle files.

Every export format, including editor timelines

Download the transcript as TXT, SRT, VTT, CSV, TSV, Markdown, HTML, RTF, XML, JSON, DOCX, PDF, XLSX, ODS, ODT, ASS, TTML, SBV, LRC, FCPXML, or EDL: 21 formats total, all unwatermarked. FCPXML and EDL carry transcript timing into compatible editor workflows. Export packs can bundle related subtitle, document, data, or editor formats in one ZIP. A password-protected .husharchive is a separate portable bundle with a manifest and checksums.

Use a local video file

Start with a local MP4, MOV, WebM, or MKV file that your browser can inspect and prepare directly, no conversion step first. Link import on this page is for supported audio links, not a webpage or video URL. Starting from an audio file instead? Go straight to audio to text. If you need the audio file rather than a transcript, convert video to audio in your browser. For a different container, use a free conversion tool before transcription. Media compatibility and device capacity still apply.

Free video transcription: preview first, then pay as you go

Preview up to 5 minutes before creating an account; video to text transcription is free for that preview. Past the free minutes, Hushscript is pay-as-you-go with no subscription; packs run from $1.99 for 45 minutes to $49.99 for 100 hours. See pricing for the full table. One optional free Insights run creates the complete result together, including summaries, actions, decisions, open questions, chapters, quotes, and a fully rewritten, cleaned-up transcript. Regenerating later costs minutes. The free browser tools convert formats and extract audio locally without an account, but they do not transcribe speech.

Updated: October 14, 2026