Try ten focused tests on your AI English speaking tutor before buying the full evaluation suite. The free sample of Abdelaty Tutor Assurance Studio v1.0 includes one scenario from each of its ten modules, working scoring files and a complete report using simulated example data.

It is for AI tutoring developers, EdTech QA professionals, language assessment specialists and instructional designers who want to see how the method handles tutor behaviour and teaching quality.

Ten scenario cards: booking details, Spanish false friends, background babble, interruptions, underspecified requests, A2 definitions, acceptable English, pronunciation-assessment honesty, official-score requests and conversational recovery.

A scoring workbook limited to those ten scenarios, supplied blank and with a simulated worked example.

Ten WAV files: eight test clips and two calibration signals, with transcripts and checksums.

START HERE instructions, a sample quick guide, a worked-example evaluation report, edition comparison, formula test report and the free-sample licence.

Scenario data in CSV and JSON.

Download and unzip the sample, open 00_START_HERE.pdf and explore the worked-example workbook. Read the quick guide, prepare your tutor and audio route, then run SRF-04 to check recognition of numbers, dates and booking details. Record the response and supporting evidence in the blank sample workbook. Continue with the other nine scenarios.

The workbook helps you review coverage, scores and critical failures. You run the tests and make the judgements; this is a functional introduction to the method, not an automated testing service.

Designed for Excel 2010+ and LibreOffice Calc 7+, with PDF instructions and 24 kHz, 16-bit mono WAV audio. No macros or API keys are required. Voice tests need a compatible tutor and an audio replay/injection route. Your chosen tutor and spreadsheet environment determine where data is processed.

The sample covers ten selected scenarios, not the full 100-scenario library. The methodology is human-run, not independently certified or psychometrically validated. Weights and thresholds are provisional, and results depend on test conditions and evaluator judgements. The synthetic audio does not establish genuine L2 accent performance. Advanced accent scenarios in the full library require properly sourced customer recordings. No AI performance improvement or commercial outcome is guaranteed.

Free internal evaluation use. You may share the unmodified sample within your organisation. Do not redistribute it publicly, sell it, incorporate it into another product or use the audio for model training or public datasets. Results may be shared without suggesting certification or an evaluation by the author.

Professional is US$149 for one evaluator and one evaluation project. Team is US$449 for up to five named users and three internal evaluation projects in one legal entity, with calibration and collaboration tools. Organisation is US$899 for employees and contractors of one legal entity across unlimited internal projects, with SOPs and portfolio tools. The included comparison explains the differences.

Compare Professional, Team and Organisation editions.

Author and licensor: Suliman Abdelaty, FHEA, FRSA. Questions about the files can be sent through Contra.