How can speech recognition technology support children’s learning?

The last 12 months have changed the world in a way that we could never have predicted, and no one can attest to this more than schools. While it was expected that e-learning would play an important role in education in the future, it was completely unexpected that it would replace classroom teaching across the world so soon.

Because of this, video conferencing has become an essential tool for teaching, whether delivered through language apps,  virtual tutoring,  video conferencing  tools, or  online learning software. It offered a portal into the virtual classroom as students were forced to stay at home. With the huge spike in both popularity and need, we could say that Zoom, and other video conference tools, were ‘made’ for the pandemic. However, as teachers and students remained at a distance, connected only through screens, delivering lessons over video brought with it its own challenges and issues – especially around accessibility and children’s ability to engage and learn via this new medium.

Optimising video with speech recognition

One technology that helps to address these issues and plays an increasing role in learning is speech technology. This is because with any device, whether it’s a mobile, tablet, or a computer used for remote learning, reading what’s being said is a major part of engagement for the students. The double learning offered through both reading and listening via closed captions is a great benefit to children.

Ultimately, speech recognition technology can optimise video platforms by enhancing the accessibility of education through captioning in real-time, for all students, including those who are deaf and hard of hearing, thus making online learning more inclusive. There is of course also the side benefit to teachers, who have a written transcript readily available after the lesson from the speech-to-text software.

Writing with voices

Not only this, students whose first language is not English may find it easier to read captions and transcripts than to follow the speech of the teacher who may have a certain accent or dialect. It’s also a great tool for kids who struggle with handwriting or spelling. For children with dyslexia and other learning difficulties, they can use speech-to-text technology to write with their voices, instead of typing or writing by hand.

Fundamentally, audio-captioning ensures that interactions in the virtual classroom and lessons can be understood beyond just verbal conversations. Teachers and students can look at the transcript at the end of a lesson which can be used as an additional learning tool or for notes to help extract as much value as possible from these online conversations. Children who were unable to attend a lesson can read through the transcripts to get up to speed.

Fundamentally, audio-captioning ensures that interactions in the virtual classroom and lessons can be understood beyond just verbal conversations

These benefits have not gone unnoticed by the industry. Further proof of the power of voice in e-learning surfaces in recent research, where 56% of respondents said that the education industry will increase its use and application of voice technology in the next three to five years.

Having said this, adoption isn’t without concerns. According to Speechmatics’ Trends and Predictions report 2021, respondents outlined some risks for voice technology in the coming years. The top concerns are linked to data privacy and compliance, poor experience and the technology falling short of accuracy expectations.

Especially when children’s data is involved, there are a lot of safeguarding and privacy concerns at play. So, when it comes to speech-to-text, there needs to be considerations linked to where the data is stored and how it’s processed. For example, using solutions which take away access to third parties are a good way to ensure schools are in control of the data.

Overcoming industry challenges

The industry also creates a unique challenge for automatic speech recognition (ASR) providers to guarantee their engines are extensive enough to recognise certain words that are used in educational settings. This remains a major challenge to ensure accuracy and the usability of transcripts as educational use cases continue to change and adapt. Having the option to include key terms which algorithms can recognise, along with punctuation, capitalisation and speaker diarisation, means that not only are the transcripts more accurate, but teachers also don’t need to filter through streams of written content, and can more easily monitor and review classroom interactions.

The pandemic has accelerated the digital transformation of a lot of industries – education arguably more so than many. Video conferencing tools with the use of ASR technology pose a great benefit for education. With such a powerful potential to have more students feel included and engaged while virtually learning, it’s no wonder respondents from recent research see the huge opportunity for speech technologies in the education industry and expect to see these develop into the future.

---

You might also like: How to plug the UK’s digital skills gap

---