Next Gen Assessments

SoapBox Fluency assesses reading fluency in kids. It can be used in noisy classroom environments or remotely and works accurately for children of all accents and dialects, generating rich and granular voice data to populate 3rd party teacher dashboards.

Clients like Amplify use SoapBox Fluency to voice-enable their reading assessment solutions because it offers a level of accuracy comparable to that of human assessors.

How It Works

Get under the hood of SoapBox Fluency and learn about the unique reading assessment data it generates for teachers with our colleague, Computational Linguist Agape Deng.

Scoring Reading Assessments

SoapBox Fluency returns the confidence scores with which each word was decoded, as well as the start and end time of each word in the utterance. 

If we click on a substitution, for example, our voice engine indicates what word the child said instead, as well as the start and end times of that segment.

Custom Language Models

Custom Language Models (CLMs) are language models that focus on a specific language domain, as diverse as fairy tales and scientific texts!

SoapBox’s CLMs are trained on kid-centric data to allow them to understand unique words and phrases with exceptional accuracy and ensure the experience of reading remains engaging, educational, and enjoyable for kids.

Abstract image depicting custom language models

The Voice Data Generated

Voice-enabled reading assessment tools generate an immediate and unique set of valuable data for students, teachers, schools and school districts.

Longitudinally, this data becomes an invaluable benchmarking tool for districts and for edtech companies, who can use it to track the performance of their literacy and language solutions in the market.

It's Time to Experiment

Whether you’re building an online or offline voice-enabled learning experience for kids, the best place to start is our Developer Portal.

Share this page
Share on facebook
Share on twitter
Share on linkedin
Share on email