Man invents ingenious device allowing you to tune out of conference calls ... and get away with it
It’s the stuff of an office worker’s dreams: a machine that allows you to entirely tune out of tedious conference phone calls while still seeming to pay attention.
And thanks to one computer genius, it may have just become a reality.
Josh Newlan, a 31-year-old Californian living in Shanghai, has written a piece of software called ‘Say What’, which lets him know whenever his name is said in the conference.
Listening via his computer’s microphone, Newlan’s device combines an open source speech recognition program and an IBM transcription server to take minutes during the meetings. Then, whenever ‘Josh’ is said, Newlan receives a transcript of the 30 seconds prior and few seconds after his name was called.
At that point, Newlan has 15 seconds’ silence to catch up on what he is being asked for input on, before a pre-recorded message plays of his voice saying, “Sorry, I didn’t realise my microphone was on mute.”
“I end up having lots of early [and] late calls with the US office, hence the need for this tool,” Newlan, who works for data company Splunk, told CNN.
While not a programmer or formally trained in writing code, Newlan completed his project during a ‘hack week’, themed learning workshops put on by his employer.
"I conceived of this idea just last week and was able to implement it in about a day," he said. "I do run the risk of losing credibility as to whether I'm actually listening in meetings now, but I'm not too concerned about that. I made this as a joke, and my coworkers know that.”
While ostensibly a joke, Newlan’s mischievous invention shows the enormous possibilities of machine learning, a rapidly-developing field of computer science that has made the news this year thanks to various ‘chatbots’ – namely Microsoft's late drug-addled, racist assistant, Tay.
As that particular case showed, machine learning still has foibles that can lead to distinctly awkward responses, and Newlan’s ‘Say What’ device is no different. The accuracy, he says, leaves “much to be desired” at the moment, but it’s still early days.
"My meetings at the office take place mostly in Chinese anyway, so I'll need to find a Chinese speech-to-text service before I can fully check out,"