Your rating scale or codebook, applied to transcripts by an AI you can measure.
Bring the instrument your study already uses—MITI, a CTRS-style CBT rubric, HAM-D, PANSS or TLC items, or your own codebook. Tagaroo’s agent applies it to interview and therapy transcripts with the evidence span and a rationale for every rating, and you check its agreement with your human raters before relying on it. Recordings and images work too.
No sales call. Start in the browser—your trial data never leaves it.
Why Tagaroo
An AI rater that takes the first pass and shows its evidence.
Joey takes the first pass
The agent tags what it's confident about and flags the edge cases—every judgment carrying its evidence span, a severity and a written rationale. You're the expert; it does the grunt work.
Ready in a few clicks
No setup marathon. Paste or import a transcript, pick a scale from the library or upload your codebook as a PDF, and run the first pass.
Transcripts first, recordings too
Interview and therapy transcripts are the core. Audio and video are transcribed with speaker labels, and images sit in the same workspace when a study needs them.
Measured, not asserted
Score the agent against your human raters with real inter-rater statistics. Your data stays yours, and every annotation records who made it.
See how we measure itScale Library
Start from a validated scale.
Forty-one curated clinical and language coding instruments—301 tagged domains across formal thought disorder, depression, emotion, argumentation and more—each with severity anchors and curated agent skills. Search one, apply it, and start annotating in minutes.
Built for
Six jobs, in their own terms.
CBT and MI trainers, clinical supervisors, counselling programmes
Supervision & training
Upload a recorded session, apply MITI or your own competence rubric, let the agent take the first pass, then review its calls span by span—and hand the trainee a report that shows exactly which utterance earned which code.
Medical, nursing, and PA educators; simulation centre leads
OSCE & clinical skills
Apply a validated communication rubric to a recorded standardized-patient encounter, get an AI first pass with the exact utterance behind every judgment, and give the student feedback that points at what they actually said.
Psychotherapy process researchers and lab coding teams
Psychotherapy process research
Apply an observer-rated process measure to therapy transcripts, let the agent draft every code with the utterance behind it, and measure how closely it matches your trained coders before it touches a single analysis.
MI researchers, trial fidelity monitors, and MI coding labs
MI trials & research
Apply the MITI 4.2.1 behaviour counts to every session a trial records, not the few your coders have time for—with the agent drafting each code, your coders checking it, and agreement you can put in the fidelity section.
Clinical researchers, trial rating teams, and computational psychiatry labs
Clinical ratings from interviews
Apply a clinician-rated scale to an interview transcript, get an item-by-item first pass with the patient's own words attached to every rating, and measure how far it sits from your trained raters. For research and rater QA—not for clinical decisions.
Qualitative researchers, PhD students, and coding teams using a codebook
Qualitative coding & IRR
Apply a deductive codebook to interview transcripts, let the agent draft every code with the excerpt and the reason, and measure its agreement with your human coders—so the AI is a coder you report on, not a black box you cite.
The 90-second tour
See Tagaroo take the first pass.
Drop in data, pick a scale, and watch Joey draft annotations—each with a severity rating, a written rationale, and its reasoning shown live—while you stay in control. Text, images, multi-rater studies, and your own scales, end to end.
1:34 · sound on
Pricing
Start free. Upgrade when a study outgrows it.
A permanent free plan, then flat subscriptions from $29/month with the AI agent included—while the desktop QDA suites charge hundreds per seat per year and sell AI as an add-on. Live prices on the site, no sales call, and the people you invite never pay.
Frequently asked questions
What is Tagaroo?
Tagaroo is a web-based AI rater for interview and therapy transcripts. You choose a clinical rating scale from the library or upload your own codebook, an agent takes the first pass with an evidence span and rationale for every rating, and you review it and measure its agreement with your human raters. Recordings, free text, and images work in the same workspace.
How does AI-guided annotation work?
You pick or upload a scale, and Tagaroo's agent reads your data and proposes annotations: labeled spans with a severity and a written rationale anchored to the evidence. You then accept, edit, or reject each one, so the final dataset is human-reviewed and every rating stays auditable back to the text it came from.
What kinds of data can I annotate?
Text of any kind—clinical and research interview transcripts, therapy sessions, support conversations, open survey answers, documents—plus images. Transcripts can be pasted as plain text, imported as JSON, or produced from audio and video files by the built-in transcription.
Which rating scales and coding schemes are included?
The Scale Library ships 25 validated instruments with items, severity anchors, and source citations—including PHQ-9, GAD-7, HAM-D, MADRS, YMRS, BPRS, and coding systems like MITI, NICHD, and Ekman emotion labels. You can also upload any scale, codebook, or annotation guideline and Tagaroo extracts it into a live workflow.
Can Tagaroo compute inter-rater reliability?
Yes. Multi-annotator studies report agreement statistics, and a compare-and-consolidate view shows exactly where raters—human or agent—disagree, span by span. There is also a set of free browser-based reliability calculators (Cohen's kappa, Krippendorff's alpha, Gwet's AC1, ICC and more) under Materials that work without an account.
Is my data private?
The no-signup trial runs entirely in your browser—trial data never leaves your device. Workspace data is hosted in the EU, processed under GDPR, and analytics only run with your consent. Identifiable clinical data is not permitted; a free in-browser de-identifier is provided for transcripts.
How much does Tagaroo cost?
Tagaroo has a permanent free plan—no card required, with starter agent credits. Paid plans are flat monthly subscriptions that include agent credits, and the people you invite to a study never pay. All prices are public on the pricing page with no sales call, students and academic staff get 50% off Pro for their first year, and one-off credit packs never expire.
Rate one transcript, then check it against yourself.
Bring a de-identified transcript and the scale or codebook you already use. The agent takes the first pass; you compare it with your own ratings. Free to start, no setup required.