The company’s fastest growing business line has more than doubled since 2023
We are proud to officially reveal Speech Technology, a rapidly growing audio datasets service for AI tools and consumer tech products. Since 2011, under non-disclosure agreements, our teams have provided Speech Technology to some of the world’s largest and most valuable tech companies. Working with human voice talent, our own Side engineers build source data – commonly referred to as datasets – for mobile AI assistants and consumer technology products. Over the last 15 years, we have recorded more than 2.5 years’ worth of data across more than 45 languages.
Speech Technology works by creating a catalogue of studio-quality recorded words and phrases for the AI to draw from. For example, when an AI assistant responds to a user’s question, it will collate recordings of individual words to generate an answer with natural language. While the service has been around in many forms for years, an increase in interactive products and services has led to additional demand. With 39 billion Internet of Things devices expected to be online by 2030, it will only continue to grow.
Side CEO, Deborah Kirkham, said:
Quotes
“We’ve created audio for blockbuster games for years, but the skillset and practices for Speech Technology are highly specialized; we’ve been working on this quietly since 2011, and I’m so glad we have the opportunity to talk about it now. This segment of the business has more than doubled since 2023 and that direction looks set to continue – the more interactive we want our products and games to be, the bigger a role speech technology will play in the development cycle.”
Side’s Senior Director of Audio, Chris Donnelly, commented: “Creating this service line from scratch with a seasoned team here at Side has been a rare pleasure. Speech Tech and traditional audio services aren’t interchangeable, and the talent we’ve grown in-house is just as bespoke as the projects themselves. The work we are delivering at the moment is being used in literally millions of consumer devices around the planet, and I can only see the need for high-quality audio data growing in the future.”