“Amazon haes hud another stoatin’ idea” is what you might be saying right now if you speak English with a Scottish accent, because if you have a particularly heavy accent that means people sometimes find it hard to understand you, Amazon’s latest stoatin’ idea could help.
The company has just patented an audio system that can detect the accent of a speaker and translate it into the accent of the listener, which could have major implications not only for voice-activated technology like Amazon's Echo range, but also for how we interact with each other globally.
The technology is designed to enable people with different accents to communicate with each other more easily. The patent describes a bank of stored accents, against which the speaker’s voice is compared and matched. Once the accent is correctly identified, the translator converts the speech into the accent of the listener in real time, allowing the conversation to flow smoothly.
Now you’re talking my language
Real-time translation apps like Google Translate are already used regularly by around 500 million people per month, with recent updates including image translations and real-time voice-to-text translation.
The new patent from Amazon could prove useful for the millions of people worldwide who regularly speak a second or third language, but who might struggle when communicating with someone who speaks with a different accent.
Many users are still reporting problems with voice-activated assistants like the Amazon Echo, Google Home, and Apple Homepod recognizing strong accents, so Amazon’s patent could well be an attempt to address such bugs.
However, a patent application is no guarantee the new technology will ultimately be rolled out. It could be that Amazon has filed a ‘defensive’ patent to stop competitors from getting there first, or is simply carrying out research. That said, better voice activation technology for people with strong accents would be absolutely 'barrie'*.