Batch upload
Drop dozens of audio and video files at once. Each is transcribed with speaker labels on, and files you have already added are flagged.
Add every recording from a study, a client or a team in one go. Each comes back with speakers, a summary and its quotes, and the project keeps them side by side.
A semi-structured interview in which Speaker 1 asks the questions and Speaker 2 describes combining fully remote work with caring for a parent at home. It covers daily routines, keeping caring private at work, support at work, and life at home.
Speaker 2 works as a payroll coordinator for a logistics company, fully remote since 2021, and lives with their mother, who had a stroke about three years ago and needs help with meals, medication and getting to appointments. A brother takes the mother every other Saturday.
Drop dozens of audio and video files at once. Each is transcribed with speaker labels on, and files you have already added are flagged.
Give each voice a name, or pick a role such as Interviewer, Participant or Chair. The app never decides who is who.
Interviews get themes, quotes and claims to check. Meetings get decisions, open questions and action items with owners.
Files named with a participant code (P03, Participant 8) get a research summary: participant context, key themes, verbatim quotes with the speaker label, and first-pass codes for you to review.
Themes, quotes, codes and action items from every recording, in the order of the date in each file name. Action items and codes also download as CSV.
Transcripts with speakers and timestamps in formats your analysis software imports, plus plain text and SRT subtitles.
Audio and video files from your computer or phone, such as m4a, mp3, wav or mp4, including Zoom and WhatsApp recordings. This app takes files, not links.
Yes, long recordings of several hours each are fine. Every file is transcribed on its own, so one project can hold a whole round of fieldwork or months of meetings.
The transcript separates voices as Speaker 1, Speaker 2 and so on. You give them names or roles (Interviewer, Participant, Moderator, Chair, Guest); the app does not guess who anyone is. Overlapping voices can be split wrongly, so skim the labels before you rely on them.
Into your own txscribe account, like any other upload there. txscribe does not use recordings or transcripts to train models, and you can delete a recording or a project at any time. If your study has an ethics protocol, check that it allows a cloud transcription service.
It uses the minutes on your txscribe account, on the plan you already have. You start with free minutes, so you can try a few recordings first.
Yes. Each transcript exports in formats for NVivo, MAXQDA and ATLAS.ti, as well as Word, plain text and SRT, with speaker names and timestamps.
It does not analyse across interviews, count codes or build a codebook: project pages put each recording’s own summary side by side. Summaries are written by an AI model and can miss or misread things, so check quotes against the recording before you cite them.