Transcription FAQ

Hushscript is private transcription for recorded audio and video. Your source video stays on your device, Hushscript sends only prepared audio for transcription and removes it after processing, and you get an editable transcript with optional speaker labels. Correct it, add optional Insights, and export it.

No. Hushscript runs in your web browser, so there's nothing to download or install.

Yes. Preview up to 5 minutes with speaker labels before signing in, from a local file, a supported audio link, or the browser recorder. Full transcription needs an account and available minutes; eligible personal accounts get 30 free minutes once, after card verification or with a first purchase.

Choose an audio or video file for a free preview, then sign in and select the file again in the app. A new personal account's first upload gets its own free preview of up to 5 minutes before you buy minutes. Set your language and other options, start the job, then review and export the transcript. Speaker labels are free: turn on speaker detection and it adds no minutes.

No. Hushscript works from recordings you upload: record however you like, then upload the file when you choose. No bot joins your calls, and Hushscript never asks for calendar access.

Review and edit your transcript, rename speakers, or export it. You can also generate optional Insights: the first run per transcript is free, and later regeneration uses minutes. Insights include summaries and a complete Cleaned transcript.

You can transcribe recorded interviews, meetings, podcasts, lectures, voice memos, and more in supported audio or video formats. Recordings can run up to 10 hours. Hushscript prepares large files in your browser, so your device's capacity matters. You can merge related recordings before starting.

Yes, and your video never uploads. Hushscript prepares its audio track in your browser and sends only the audio needed for transcription. Compatible audio passes through without re-encoding; conversion runs locally only when required.

You can import a supported audio recording from a shareable link. Video links, including YouTube video links, cannot be imported; to transcribe video, use a local file so Hushscript can extract its audio on your device. Only submit recordings you have the right to use.

Yes. Recordings must be at least 15 seconds long and can run up to 10 hours. A shorter recording is rejected before transcription is charged.

Processing time varies with recording length, format, and service load, and Hushscript does not guarantee a turnaround time.

If someone follows your referral link before creating an eligible new account, their first eligible paid pack gets 30 to 300 bonus minutes depending on the pack. You may earn minutes on eligible purchases during their first 60 days. See the Referral Program for the conditions.

Yes. Completed purchases appear in Billing with a stable receipt number and a downloadable Hushscript payment receipt. It confirms the recorded payment and is not a tax invoice.

A preview of up to 5 minutes is free before sign-in, and a new personal account gets one more free preview of up to 5 minutes from its first upload in the app. Eligible personal accounts can receive 30 free transcription minutes once after successful card verification, or with their first purchase if not already granted; further transcription uses prepaid packs. The in-browser tools are free. Teams has separate paid plans.

Hushscript gives eligible personal accounts 30 free minutes once. To claim them, verify a card: Hushscript authorizes a temporary $1 hold and releases it without charging, though your bank may take time to remove the hold. Your first purchase also grants the minutes if they have not been granted already.

No card is needed for the preview. For the 30 free minutes, card verification is one route and your first purchase is the other. Packs use a payment method shown at checkout.

Personal transcription uses five prepaid packs, from $1.99 for 45 minutes to $49.99 for 6,000 minutes. See the Pricing page for the full pack list and effective rates. Teams plans are separate.

Personal packs range from about $0.044 per minute on the 45-minute pack to about $0.008 on the 6,000-minute pack. You buy the full pack up front; medical mode, translation targets, and later Insights regeneration can use additional minutes.

Personal accounts use prepaid minute packs without a subscription. Hushscript for Teams is a separate monthly or annual subscription with pooled hours.

Card verification confirms eligibility for the one-time 30 free minutes and keeps that credit from being abused. Signing up is free.

No. Signing up is free and there is no subscription: you pay only for the minute packs you choose to buy. Claiming the welcome minutes through a card check carries no charge.

No. Minutes do not expire.

Transcription uses minutes according to the recording's duration, including silent stretches. Medical mode adds 100% of that duration, and each selected translation target adds 25%. With multichannel on, each audio channel is billed separately, so a stereo file counts twice. Trim dead air before uploading if you want to use fewer minutes.

No. Base transcription is billed by recording duration, whether your source is audio or video. Your video stays on your device while Hushscript sends only its prepared audio for transcription. Optional medical mode or translations add minutes.

The first run is free and generates everything together, including a fully cleaned transcript. Regenerating later costs minutes: 40% of the recording's length to start over from the Original, 30% to refresh from your Cleaned version. The exact cost shows before you confirm.

You're not charged. The minutes held for a job are released back to your balance automatically if it fails on Hushscript's side.

No. Minutes are delivered to your balance the moment payment succeeds, so the sale is complete at purchase and all sales are final and non-refundable. Failed transcriptions return their minutes automatically, and genuine billing errors are always corrected: email support@hushscript.com. Where the law that applies to you grants a non-waivable refund right, that still stands.

Stripe processes payments using the methods shown at checkout. Hushscript never receives or stores your full card number.

No. The in-browser tools are genuinely free, with no signup.

The first Insights generation per transcript is free. That one free run can produce summaries, actions, decisions, open questions, quotes, chapters, speaker topics, subject clusters, export suggestions, terminology and profanity findings, and a fully rewritten Cleaned transcript. Later regeneration uses minutes.

Cleanup is part of the free Insights result: a fully rewritten Cleaned version of your transcript, not a list to approve line by line. It lives in the Cleaned tab, where you can compare it to the Original, edit it, or revert anytime.

No. Your source video never leaves your device, and Hushscript removes its transcription-audio copy after processing. It also asks the speech provider to delete its transcript copy once the job is finalized; if that request fails, removal can take longer.

No. Your video never leaves your device. Hushscript copies a compatible audio track out in your browser without re-encoding, converts locally only when required, and sends only that audio.

You decide. For personal accounts in normal mode, keep saved transcripts until you delete them or set auto-deletion after 7, 30, 90, or 365 days. You can override the default for one transcript. Teams workspaces have their own retention rules, and a legal hold can prevent deletion.

Private mode saves no transcript to your account. Speech recognition still runs online, and Hushscript returns a password-protected .husharchive instead and skips post-processing Insights. Medical mode and translation aren't available in private mode.

Yes. Delete a transcript through its dedicated flow after confirming the action. That flow returns a downloadable JSON deletion receipt. Scheduled and bulk deletions store receipts but do not hand you a download at the time.

Yes. Delete it yourself in your profile, after a confirmation step. Any minutes you've bought aren't refunded when you delete the account.

No. Hushscript never uses your recordings to train models. Prepared audio goes to a speech-recognition provider for transcription, and Hushscript removes its own transcription-audio copy after processing.

Yes. Hushscript encrypts stored transcripts, titles, speaker names, translations, and Insights at rest, and uploads and downloads use HTTPS. No service can promise a breach is impossible, so Hushscript keeps less: no transcription audio after processing, and retention you control.

Audio from users in the EU, EEA, Switzerland, and the UK is processed in the EU. Teams workspaces are pinned to one of four data regions: US, EU, Asia-Pacific, or Oceania. Check the Privacy Policy for the applicable processing and storage details.

A third-party speech-recognition provider processes the prepared audio and returns your transcript. Hushscript removes its own audio copy after processing, asks the provider to delete its transcript copy, and never uses your recordings to train models.

Hushscript is operated by a Hong Kong company. EU, EEA, Swiss, and UK audio is processed in the EU. You can request access, correction, or deletion via support@hushscript.com. Contact the same address with questions about a data processing agreement. See the Privacy Policy for how your rights apply.

The public site uses Google Analytics and Google Ads, subject to applicable consent. The public site may also load advertising tags, including Meta Pixel, subject to your advertising consent settings. Server-side Meta Conversions API reports selected events where permitted. The app loads no analytics or ad scripts and sends no browser product telemetry. Hushscript's API reports content-free signup, payment-method, checkout, purchase, and selected product milestones. Hushscript does not send transcript text, filenames, prompts, questions, answers, or form contents to analytics. See the Cookie Policy.

Yes, that is what Hushscript is built for. Your source video never leaves your device, Hushscript removes its transcription audio after processing, saved transcripts are encrypted at rest, and you set how long anything stays. Check names and numbers before professional use, and make sure you have the right to record and transcribe the material. One limit: Hushscript is not for protected health information under HIPAA, as the next answer explains.

No. Hushscript itself does not sign HIPAA Business Associate Agreements (BAAs) with customers and is not a HIPAA business associate, so covered entities should not rely on it for protected health information under HIPAA. Our AI processors are covered: AssemblyAI (transcription) processes audio and transcripts, and OpenAI (Insights, cleanup and translation) processes transcript text, for us under signed HIPAA BAAs. We have no BAA with our storage, hosting and diagnostics providers, including Cloudflare, Sentry and Mailgun.

Dedicated transcript, AI artifact, and account deletion flows return a downloadable JSON receipt recording what was removed and when.

Accuracy depends on the recording, language, noise, and overlapping speech. Clear audio usually helps, but names, numbers, technical terms, and speaker labels can be wrong. Review and correct the transcript before relying on it.

Use the clearest recording you have: close mic, minimal background noise, one person speaking at a time. Converting or compressing a noisy file won't add detail that isn't there.

Yes. Before upload, apply a saved dictionary for names, abbreviations, brands, drug names, and domain terms; apply a saved prompt preset for transcription behavior; then add one-off keyterms or a one-off custom prompt for the specific recording. These inputs affect transcription, not Insights.

Transcripts are generated by machine and can contain mistakes, so review them before relying on them, especially names, numbers, and technical terms.

Yes. Edit transcript segments in Hushscript, split or merge segments, reassign speakers, undo and redo changes, find and replace text, rename speakers, add tags, and revert to the original transcript if needed.

Yes. Timestamps are included; export as SRT for caption-ready lines.

Hushscript adds automatic speaker labels at no extra minute charge when you select speaker detection. Labels can be wrong, especially with overlapping speech, so check them in the editor.

Hushscript labels multiple speakers automatically, with no paid speaker cap. Separation quality depends on the recording.

Yes. Change 'Speaker 1' to a real name in one click, across the whole transcript.

No. Speaker detection and labels add no minutes.

Hushscript supports roughly 99 spoken transcription languages and detects the source language automatically. Language support does not guarantee the same accuracy for every recording. See the Languages page for the list.

No. Hushscript detects the language automatically, and in the full app you can set the source language yourself instead. The Languages page has the full list.

A recording can contain more than one supported language, but language changes can affect recognition quality. Review the transcript and correct any missed words.

Yes. Choose translation targets before upload. Translated transcripts appear as language tabs and can be exported per language.

13 today: English, Spanish, French, German, Italian, Portuguese, Dutch, Hindi, Japanese, Mandarin Chinese, Thai, Turkish, and Vietnamese, all chosen before you start. That's separate from the roughly 99 languages Hushscript transcribes automatically: the source is whatever's spoken in your recording, and the 13 are the targets you pick for the output.

Yes. Each translation target adds 25% of the recording duration to the minutes used. Select targets before starting. A job with translation targets, including its base transcription, uses your personal minute balance even if the transcript belongs to a Teams workspace.

Common audio and video formats include MP3, M4A, WAV, AAC, FLAC, OGG, MP4, MOV, WebM, MKV, and more. Compatible audio uploads as-is. Video audio is copied out without re-encoding when possible, with local conversion only when required.

Contact support@hushscript.com with the file extension and a description of the recording. You can also use the free browser tools to convert some files to a supported format; conversion depends on your device and the source file.

A recording can run up to 10 hours. There is no fixed source file-size limit, but large files are prepared in your browser and depend on your device's capacity. Service-safety and active-job limits also apply.

Export transcripts in 21 formats, including TXT, SRT, VTT, CSV, TSV, Markdown, HTML, RTF, XML, JSON, DOCX, PDF, XLSX, ODS, ODT, ASS, TTML, SBV, LRC, FCPXML, and EDL. Format packs group related downloads. A password-protected .husharchive is a separate archive option; Insights can also add show-notes and meeting-minutes exports.

Yes. Export SRT for timestamped, caption-ready lines. Already have a subtitle file and only need a format swap? Use the free SRT↔VTT converter.

No. Every export format is included at no extra charge.

Included. SRT, VTT, ASS, TTML, SBV, and LRC, plus FCPXML and EDL for editors, are part of the same 21 export formats every transcript includes, not a separate product or a paid add-on.

Speaker labels can appear in subtitle exports when speaker detection was used. Check the labels and timing before publishing captions; automatic speaker attribution can be wrong.

Yes. They run entirely in your browser, with no signup and nothing uploaded.

No. Everything happens on your device; your files never leave it.

Convert between formats, extract audio from video, compress audio, and more. See the free tools page.

They run entirely in your browser, so the practical size limit depends on your device's memory rather than a fixed cap.

The free tools run on your device and do not require an account. They process your selected files in the browser, so Hushscript does not receive those files for tool conversion.

With a one-time link sent to your email, or with an email and password.

Check your spam folder; if it still hasn't arrived, contact support@hushscript.com.

There's no app to install. Hushscript runs in your web browser, on desktop, tablet, or phone.

Yes. The REST API uses a personal access token. Hushscript also offers an anonymous documentation MCP server and an account MCP server that accepts OAuth 2.1 or a personal access token. Account actions follow the applicable personal-balance or Teams-workspace funding rules. See hushscript.com/developers and hushscript.com/mcp.

Yes. Hushscript for Teams offers shared workspaces and pooled hours on separate monthly or annual plans. Self-hosted Hushscript is not offered. See the Hushscript for Teams page for current plans and workspace controls.

Email support@hushscript.com for general questions, support, billing, and privacy/DPA. We respond within 1–2 business days.

Hushscript is operated by LW Agency Limited, a Hong Kong–incorporated company. See the About page.