A Malaysian programmer was inspired to devise his own real-time translator after he experienced trouble understanding his Mandarin-speaking colleagues during meetings.
Twitter user Roshen Maghhan (@roshenmaghhan) recently shared his DIY project which took inspiration from Google Pixel Buds, a type of wireless earbuds that can convert any language in real-time.
In a series of tweets he posted on Tuesday, Maghhan wrote how he’d usually end up confused after their casual meetings in the morning at a local software development company as most of his coworkers would converse in Mandarin.
Despite the amount of times I’ve I asked them to converse in English, they WOULD acknowledge it, but eventually start conversing in Mandarin again after a while, & I believe its unintentional, bcs they’re very nice
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
“About six months ago, I worked at this software development company, and every morning, we’d have a casual meeting about the work,” he wrote. “But half the time, during the meeting, most of them would converse in Mandarin, and so I’d always be pretty clueless half the time.”
He lamented that when he would ask his colleagues to converse in English so he could understand, they would acknowledge it initially but would eventually revert back to speaking Mandarin again later on.
“Despite the amount of times I’ve I asked them to converse in English, they would acknowledge it, but eventually start conversing in Mandarin again after a while, and I believe its (sic) unintentional, [because] they’re very nice,” he explained.
“Then I remembered about the Google Pixel Buds, and how it’s probably the tech solution to bridge the gap between language barriers,” Maghhan added, posting a video showing how the technology works:
Then I remembered about the Google Pixel Buds, & how it’s probably the tech solution to bridge the gap between language barriers.
If you dont know what the Pixel Buds are, they’re able to convert any language in real time, like this : pic.twitter.com/TFmlb9WUet
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
Since a pair of the wireless earbuds would cost him almost RM800 ($200), he decided to build the device himself using Google Pixel Buds’ hardware information (obtained from the United States’ Federal Communications Commission), Google Translate API (purchased), ResponsiveVoice API (voice synthesizer), and a normal pair of earphones with a microphone.
Alright, so first, we’ve got to reverse engineer this product to see how it works, you can check out the hardware info from the FCC’s website, this is what I usually do if I wanna hack hardware devices.
Link to the FCC website : https://t.co/HIHzi50FfU pic.twitter.com/xMF9y7xD0e
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
I wanna make sure it’s able to work on every platform ( Android, IOS, Desktop ) therefore, it’ll be web-based
Also, I wanna make sure its able to work with any earphone wired or wireless, as long as it has a microphone. So here’s the earphones I’ll be using ( Left pic ) pic.twitter.com/Gbaz4XvRRs
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
If you wanna train an A.I for your translation, you’re gonna need a lotttt of training data PLUS, you’ll need to make sure your device has enough computing power, PLUS it’ll be an extremely lengthy process, so instead of re-inventing the wheel, just use Google Translate’s API
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
I dont wanna bore you guys, I’ll just jump in to the end product.
For voice synthesis, i used ResponsiveVoice API, and throughout this entire demo now, every synthesized audio you hear, will come out from my earphones pic.twitter.com/HmJpRFcKMD
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
After making it work for his intended use, he then decided to take it a step further by adding the ability to differentiate voices, known as ‘voice diarisation,’ a tool which he developed himself:
Now I realised what if I actually translate ALL speeches to text in real time, and implement voice diarisation to differentiate each voice?
This way, even deaf people can be involved in a convo, and will be able to read & know what the convo is about pic.twitter.com/DMoRzUBdAI
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
Maghhan ended his thread by sharing that his nifty gadget will now let him stay in the loop during meetings and allow him to know if someone is talking shit about him “in any language.”
Modern day problems require modern day solutions, amirite 😉
Now you can talk shit about me in any language and I’ll still understand, and I can finally day dream during a convo and still stay in the loop HAHA
— roshen maghhan part 2 (@roshenmaghhan) July 9, 2019
Featured image via Twitter/roshenmaghhan
Join the conversation+