Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Processors/DSPs

Voice, gesture controls double as always-on observers

Posted: 02 Jun 2014 ?? ?Print Version ?Bookmark and Share

Keywords:always-on? voice? Dash? Siri? Glass?

Apple's Siri is an intelligent personal assistant and knowledge navigator that leverages voice interface instead of touch to perform tasks or services on an iPhone or iPad. The Google Glass user interface, meanwhile, partly relies on voice commands for operation and performing tasks.

More recently, e-commerce giant Amazon unveiled a microphone-enabled barcode reader, the Amazon Dash, giving a helping hand to add items to your shopping list. The device was given away free of charge to a select number of Amazon's Prime Fresh loyalty program users who can now order their groceries either vocally or through a quick scan of a product's container (for example a milk bottle).

Just say it and get it delivered to your home. "Never forget an item againDash remembers so you don't have to" it says on amazonfresh's website.

More broadly, the data you generate as you speak into these voice-activated devices will be scrutinized beyond their superficial user-interface functionality. Most of the data will transit from the application to the cloud where more powerful data analytics can be performed, either to bring a better match to what you're looking for, or simply for better consumer profiling and to prompt you with useful spending reminders.

Apple has already admitted that it kept the data for much longer than you would think is reasonable for a mere user-interface (up to two years). In effect, all these voice controls double as always-on eavesdroppers, adding another layer of intrusion and control over your life.

Of course at hardware level, the audio processing engines and DSPs that support these voice commands are only enablers. But these chips are getting more powerful year-on-year while becoming more energy-efficient, hence finding their way into more electronic devices.

Wolfson recently boasted a partnership with Audyssey, set to revolutionise audio recordings on mobile devices with Audyssey's so-called AudioFrame. Running on Wolfson's WM8280 Audio Hub solution, AudioFrame correlates the audio signals of two microphones to focus on the signal coming from the front (facing the smartphone's camera) and then reduces the uncorrelated ambient signals from the side.


Figure 1: Audyssey's AudioFrame microphone zooming feature. Source Audyssey

In effect, the recorded audio signal can be dynamically adjusted to match the video zoom, making the subject in the centre of the frame sound closer.

That's more microphones directed at you, think audio crowd-sourcing. If such audio zooming capabilities can be integrated into mobiles, you can bet they will combine well with audio analytics in surveillance cameras.

1???2?Next Page?Last Page

Article Comments - Voice, gesture controls double as al...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top