I spend a lot of time keeping up with the #Microsoft technology roadmap, research and releases. It’s relentless and often a thankless task (actually, the nice people at our PPUG meetings do appreciate it)
Today I needed to review their recent article on cognitive load in online meetings for an article I’m writing; it’s long and my eyes were tired from 4 hours of almost uninterrupted VDU time, so I thought I’d have it read to me.
The new Chromium based Edge browser (which is extremely good) has a handy Read Aloud feature. I’ve used it before, but the artificial voice is always a little wearing; much better than listening to the much missed Stephen Hawkings, but not exactly soothing. Jumping into the voice settings to see if I could find one I was a bit happier with, I noticed a new set of voices. They weren’t properly grouped with the other UK (or proper English, as I remind my US friends) voices, so easy to overlook.
It’s instantly impressive; far more human sounding, less awkward and distracting, far more soothing to listen to.
Microsoft Mia Online (Natural) – English (United Kingdom) caught my eye, so I tried it. You should too. It’s instantly impressive; far more human sounding, less awkward and distracting, far more soothing to listen to.
‘Online (Natural)‘ is a big clue as to why this is different from the other voices. These voices use the Azure Cognitive Services platform and its remarkable machine learning capabilities to press the text to voice in the cloud rather than locally. I had come across this development last year while at Microsoft Ignite; but had let it slip my mind as it wasn’t available in the UK at that point; trying it for real brought it all flooding back. The promise then was a lifelike speech with the natural-sounding speech patterns and intonation of a human voice rather than the robotic intonation we have come to tolerate.
It’s impressive that this technology now works and is quietly embedded in ‘business as usual’ applications like a browser. It’s even more impressive that this stuff is just another cloud-service; you can add it to your own applications at little cost without being a god-like coder.
You can read more about Azure Speech services too, if that’s your thing.
Right, I need to get back to it – there is a nice sounding young lady wanting to tell me about Microsoft Teams.