OwnVoice — Shows students which passages may read as AI-written, before they submit.
A free tool for students that shows which passages of a thesis may read as AI-generated, so they can rewrite them in their own words before submission. It gives likelihoods, never verdicts, and keeps a drafting record to show a supervisor.
The problem
Students are being flagged by AI detectors after submission, with no chance to respond, and every university sets a different policy on AI use.
What we built
A pre-submission check that flags passages with a likelihood and guides the student to rewrite them (it never rewrites for them unless their university allows declared AI use), with checks tuned to the student's stated university policy and a version history of their drafting.
What it does today
A likelihood per passage (never a verdict), guidance to rewrite in their own words, and a drafting record a student can show their supervisor.
— Under the bonnet
What makes a product like this hard to build
- 01
Likelihood, never a verdict
AI-writing detection is uncertain. The tool shows a likelihood per passage and explains it, so a student is never told they cheated by a score.
- 02
Long documents
A full thesis is tens of thousands of words. Analysing it passage by passage, quickly, and showing the results in context is a document-processing problem in its own right.
- 03
Every university has its own rules
AI-use policies differ by institution, so the checks and the guidance change with the student's stated policy.
- 04
A drafting record
Keeping versions of the student's drafting, so they can show a supervisor how the work developed, needs careful storage and privacy handling.
— Estimated build cost
What a build like this costs
£110,000–£160,000
AI writing-analysis tool · indicative range for a comparable build
Sized from the codebase we built: 112k lines of production code, 68k lines of tests, 29 database tables.
A product like OwnVoice is a bespoke build, not a template: most of the cost is in the hard parts above, plus testing against real data before anyone relies on it. Where your version lands in the range depends on the factors listed here. You get a fixed price once scope is agreed — usually after a short discovery sprint. Hosting and AI model usage are billed at cost on your own accounts.
Get an estimate for yoursWhat moves the price
- Document length and formats (Word, PDF, plain text)
- Analysis method and its accuracy
- Policy rules per institution
- Version history and storage
- Privacy and data-retention requirements
Built with
- TypeScript
- PostgreSQL
- Python
Designed, built and run by our Manchester team. Want something similar? See our AI development services, AI automation and pricing.
