Using ASR in Children’s Reading Monitoring and Development

In recent years, technology has revolutionised many aspects of education—but one critical area remains ripe for innovation: reading aloud practice. Reading aloud is a cornerstone of literacy development, yet it’s often underutilised. Classrooms are busy, one-on-one reading time is limited, and many children feel self-conscious reading aloud in front of peers or adults. So how can we empower children to read aloud more, and more confidently, while reducing pressure on schools and teachers?

One answer lies in Automatic Speech Recognition (ASR)—a form of artificial intelligence that is already powering tools like Siri, Zoom captions, and virtual assistants. But how can we harness ASR in a way that genuinely supports children’s reading development?

What Is ASR, and Why Does It Matter for Reading?

Automatic Speech Recognition refers to technology that listens to spoken language and converts it into text. Some ASR systems, like transcription software, are designed to tidy up and paraphrase speech—removing ums, ahs, and stumbles to produce neat sentences. Others, such as voice-controlled assistants, are built to pick up key words or phrases and act on them.

But these typical, commercial ASR systems aren’t designed for children, nor are they sensitive enough to the specific needs of learning readers. When it comes to reading aloud, what we need is an ASR model that behaves more like a supportive teacher—one that can distinguish between correct pronunciation, misread words, and natural variations in accent or fluency.

Reimagining Reading Aloud with ASR

Imagine a child using an EdTech application that integrates a tailored ASR system:

• The child is presented with a book or line of text.

• They read it aloud to the device.

• The ASR listens carefully, checking in real-time whether the child has pronounced each word correctly.

• Instant feedback is given—highlighting words said correctly or flagging errors.

Over time, the app tracks patterns in reading mistakes and progress.

This approach offers three major benefits:

1. For Children: More Confidence, More Fun, More Progress

ASR makes reading aloud a low-pressure, independent activity. With no need for an audience, children can practise at their own pace, in their own space, and build fluency over time. Real-time feedback helps them spot and correct errors as they go—transforming reading into a game-like challenge with scores, rewards, and motivation baked in.

Crucially, this supports what research shows: children who enjoy reading perform better. By gamifying the reading experience, ASR can boost both confidence and reading frequency—key factors in long-term literacy development.

2. For Teachers: Insight, Not Overload

Teachers can’t be everywhere at once—but ASR can help fill the gap. When children read with an ASR-enabled app, they generate rich data about their progress. Educators can then use this data to:

• Understand where students are struggling (e.g. specific phonics or decoding issues).

• Tailor 1:1 support sessions more effectively.

• Track improvement over time.

• Suggest targeted reading materials.

In short, ASR lets children get more quality practice time, while freeing teachers to focus on deeper, more personalised interventions.

3. For Developers: A Call for Purpose-Built, Inclusive Technology

Despite the potential, most off-the-shelf ASR systems are not fit for this purpose. Commercial ASR tools are typically designed for adult voices, fluent speakers, and tidy transcripts—not the messy, wonderful reality of a seven-year-old learning to read.

An effective ASR model for reading aloud must be:

• Phoneme-sensitive: able to detect subtle pronunciation differences.

• Explainable: offering transparency and control over how decisions are made.

• Inclusive: accommodating accents, fluency levels, and age-based variation.

• Lightweight and real-time: able to run efficiently on a variety of devices and provide immediate feedback.

This is no small task. It requires thoughtful collaboration between speech scientists, educators, child development experts and technologists. But the reward—a tool that truly supports every child’s reading journey—is worth the challenge.

Final Thoughts

ASR technology, when purposefully designed and embedded into reading applications, holds incredible promise for transforming how children learn to read aloud. It allows for independent practice, real-time support, and data-driven teaching—all in one. But to make this vision a reality, we must move beyond commercial “black-box” systems and invest in creating child-focused, educator-informed, and inclusive ASR solutions.

At Auris Tech, we believe the future of reading should be engaging, empowering, and equitable for every learner. ASR can help us get there—if we build it the right way.