Irish tech firm helps kids’ voices be heard
Kid-focused companies everywhere are seeing the potential for voice technology to deliver joyful and engaging learning and play experiences for children. SoapBox’s speech recognition software powers these voice experiences for kids ages 2 years and up.
Our proprietary voice recognition software for children is available as a lightweight, low-code API for fast and easy off-the-shelf integrations across a wide range of use cases, from classroom instruction and remote learning to independent play and beyond. We also build custom models for transcriptions and fluency assessments.
The most accurate voice engine for kids in the world
A world-class and diverse team of experts
A company focused on social impact and equity
AI built using privacy-by-design
Our founding story
Inspired by her oldest child and her background as a speech engineer at companies like IBM and Bell Labs, Patricia Scanlon, PhD founded SoapBox Labs in 2013 to redefine how children interact with technology using their voices.
Most speech recognition systems, including those used in your home smart speakers, are built using adult speech data and modeled on adult behavior. These systems don’t work well for kids. Patricia’s vision was to create a different kind of speech recognition technology that could accurately understand all kids’ voices regardless of accent or dialect, and that was flexible enough to work off-the-shelf or be customized for bespoke use cases. Learn more about our technology.
Patricia has been named one of the world’s top women in tech by Forbes and a “Visionary in Voice” by industry leading publication Voicebot.AI. Her 2019 TEDx Talk, “How Technology Transforms a Child’s Reading Journey,” is a must-see for professionals in the voice, education, media, and kids’ entertainment industries.
The SoapBox team has over 150 years of experience in AI and speech recognition development. Our machine learning expertise spans audio signal processing, speech analytics, speech recognition, deep learning, computational linguistics, biometrics, and multi-modal processing and recognition.
We also understand the profound implications of voice technology for kids’ data privacy and have extensive experience and understanding of data and AI regulations. Protecting kids’ fundamental right to voice data privacy is a cornerstone of our company values and philosophy.
Dr. Patricia Scanlon
Founder & Executive Chair
Dr. Martyn Farrows
Dr. Amelia Kelly
VP of Engineering
VP of Product
Do you believe in the power of accurate and safe voice technology to help kids to learn, play, and shape their worlds? Join our diverse team, remotely or here in Dublin, for a dynamic and collaborative role with some of the best deep-tech minds in the industry, and hands down, the best voice AI company for kids in the world.
SoapBox Labs is based in Dublin, Ireland. Come visit us at our city-center location sometime! We are a remote-first company, so while the vast majority of our people are based here at HQ, we also have team members across Europe, the UK, and the US.
SoapBox’s mission is to make all kids’ voices heard in the digital world. Our voice technology powers magical, joyful, and frictionless experiences for children regardless of whether they’re using online or offline devices.
SoapBox works with media, education, and entertainment companies across the world. We power these companies’ education tools and platforms (literacy, ELL/ESL, math, etc.), as well as their gaming, robots, interactive TV, and XR experiences.
Our clients range from household names like McGraw Hill and Mattel to digital pioneers like Norby and Lingumi developing voice-first experiences.
For most SoapBox customers, our off-the-shelf API, which leverages our pretrained models, can be integrated quickly and easily to power their use case. Off-the-shelf integration is available for use cases as diverse as command and control, search, utility, menus, multi-choice questions, pronunciation assessments, conversation, keyword spotting, and moderation filters.
Custom models can also be built for customers whose use cases require speech transcripts, voice chat transcripts, dictation, or reading fluency assessments.
SoapBox’s speech recognition technology is world class and leads the market for a number of reasons. Firstly, it’s proprietary and built to work accurately for kids’ voices ages 2 years and up.
Most other speech recognition systems are built primarily on adult voice data and model adult behaviors. Therefore, they don’t work well for kids. Kids are unique, and building speech recognition software for them is complex given their unpredictable speech patterns and behaviors, diverse accents and dialects, noisy environments, and their privacy needs.
SoapBox’s voice engine has been built to accommodate all of these complexities. It can accurately understand kids’ speech down to the phoneme level and has been independently validated to show no bias in speech recognition between kids of different accents and ethnic backgrounds. To learn more about this independent research please email us: email@example.com.
SoapBox’s voice engine is also unique in the industry because it gives you the power to build — and change — the voice experience you’re offering (specific brand names, non-dictionary words, etc.) without the need for customization. And when you do need a custom approach (e.g., for speech transcriptions and reading fluency assessments) we can build custom models for you.
SoapBox’s voice engine is a game-changer for education companies in particular, where identifying subtleties in a child’s pronunciation (pronouncing “color” as “other”, for example) can have a big impact on their learning progress.
SoapBox is also an independent company, whose technology is built from the ground up by our in-house team of experts. We focus only on building world-class voice technology for kids and have no business models around data, advertising, or marketing. We are a privacy-first company and proactively protect children’s fundamental right to voice data privacy.
Learn about the features and use cases of our voice tech on our Technology page.