Higher Education

An AI-Powered Reading Tutor for Every Child


Making reading assessments more efficient using voice recognition to help kids grow their love for learning so that they can build a brighter future.




Data Monsters

Use Case

Speech Recognition


NVIDIA DeepStream, NVIDIA Riva, NVIDIA T4/A40/V100 Tensor Core GPU, NVIDIA Triton Inference Server

Plabook’s mission is to help kids grow their love for learning so that they can build a brighter future. The company’s diverse portfolio of education products includes an innovative reading platform that uses speech AI to teach children how to read.

Founded in 2008, Data Monsters is an AI R&D lab and consulting company. Its scientists, engineers, product managers, and business consultants are passionate about data, science, and creativity.

A Much-Needed Resource for Teachers and Students

According to Education Week, 75 percent of all fourth graders in the U.S. read below their grade level. Early reading assessment and intervention are key to a student’s success. A running record, a method for evaluating a child’s reading level, requires more than 10 minutes of a teacher’s attention for recurring mandatory assessment. A class of 20 students equates to more than three hours of assessment, three to four times per year.

NVIDIA Solution

Plabook, an intelligent platform that helps students learn to read, automates reading assessments, such as oral reading fluency, dyslexia screening, running records, comprehension, and phonemic awareness. The app listens to a child read aloud, makes an assessment, and generates a digital running record.

Data Monsters used the NVIDIA Riva Speech AI SDK to build accurate speech capabilities into the Plabook application with voice recordings from hundreds of kids with varying accents and reading levels.

Plabook Results


  • Reduces manual assessment from hours to minutes

  • Flags and records reading errors in real time

Plabook improves student reading levels and increases literacy skills and comprehension. The platform also saves teachers time since the AI manages the recordings and highlights errors that they only need to validate. The process requires minutes, versus hours, of a teacher’s time for an entire class.

“Using NVIDIA® Riva automatic speech recognition, we’re able to assess students’ reading accuracy at the phoneme level and provide personalized feedback for improving their reading levels, saving hours of teaching time.”

Dr. Philip Hickman
CEO, Plabook