Research transcriptionTranscription built for fieldwork

Qualitative data has a bottleneck, and it isn’t collection: it’s the gap between a folder of recordings and text you can code. Research transcription with Hushscript is built for that gap – a study’s worth of interviews staged as one batch, focus groups with every voice labeled, multilingual fieldwork handled by one tool instead of a vendor per language, and exports that drop into analysis software with timecodes and speakers intact.

Research transcription in three steps

Stage the recordings and start them together; a multi-session interview can be merged into one ordered transcript. Preview the first 5 minutes of a file to check speaker separation on your actual participants, then sign up to run the corpus. Your first 30 minutes are free: add a card for a $1 check (released right away, never charged), or pay another way and they arrive with your first purchase.

Fieldwork in one language or nine

Recognition covers roughly 99 languages with automatic detection, so a comparative study doesn’t need a transcription vendor per country. Translation targets (+25% of the duration each) put the original and your working language in tabs of the same document: the participant’s exact words stay quotable while you code in the language you think in. Each language exports separately.

Focus groups, attributed

Speaker separation is automatic, free, and uncapped, which is what a six-participant focus group needs. Labels carry no identity – rename them to your participant ID scheme and the pseudonyms hold through the transcript, the exports, and anything derived from them. The honest caveat: crosstalk is the hard case, and the busy stretches of a lively group deserve a review pass with the timecodes pointing you at the right moments.

From transcript to coding

The transcript is the start of analysis, not the end. Exports cover 21 formats with a Research pack bundling the useful ones: CSV, TSV, and XLSX for spreadsheet-based coding, JSON for scripted pipelines, DOCX for the appendix. Timecodes and speaker attributions survive export, so a quote in your findings traces back to its moment in the session. Saved dictionaries hold the study’s terminology; find-and-replace fixes a mistranscribed place name across the corpus at once.

Answers for the data-management plan

Ethics review asks where the data lives, and the answers are short. The audio is deleted the moment each transcript is ready and is never used to train models. Transcripts are encrypted at rest, under retention windows you set – auto-delete after 7 to 365 days, or per-transcript overrides – with receipts from dedicated deletion flows and a JSON export of account data on demand. EU, EEA, Swiss, and UK audio is processed in the EU. For a recording that shouldn’t leave a stored transcript at all, private mode returns a password-protected archive instead. These are capabilities to check against your protocol, not a certification.

Priced for research budgets

On the largest pack ($49.99 for 100 hours, about $0.50 an hour), twenty hour-long interviews cost about $10, and a 90-minute focus group about $0.75. There’s no subscription: fieldwork-heavy months cost what they use, analysis months cost nothing, and unused minutes stay valid for a year, with any activity resetting the clock.