Determine if you have diabetes just by your voice. We are entering an era where voice can tell us a lot.
【AI determines if you have diabetes just by hearing your voice.】
・An artificial intelligence (AI) model has been developed that can diagnose diabetes in the human voice: read a sentence for 6 to 10 seconds, analyze the acoustic characteristics of each voice, and find out if you have diabetes
・A team of researchers at the Crick Institute in Canada has developed an AI that can determine if you have type 2 diabetes with more than 86% accuracy based on basic health data such as age, gender, height, and weight, and a voice recorded on a smartphone.
・The research team asked 192 healthy people and 75 people with type 2 diabetes to record one sentence each day on their smartphones six times a day for two weeks. Data on the participants’ age, gender, height, and weight were also studied. From the 18,465 recordings, the team analyzed 14 acoustic features and found differences
・The research team stated that “voice technology will bring about a significant change in diabetes decision-making methods due to its accessibility and low cost,” and that they will expand their research so that voice diagnostics can be used in areas such as diabetes, femtec, and hypertension in the future.
These are the quotes from the article
Toward an age where voice can tell us a lot.
【Turns out that having girls in the audience “enhances the vocal quality of the boys’ choir”.】
There was also this article related to voice.
「The singing voice of the boys in charge of the bass, aged 16-19, was found to be enhanced at a specific frequency (2500-3000 Hz) where sound energy is concentrated when girls aged 15-16 are in the audience」
In other words, the frequency of the singing voice indicates that the boys are straining to make their voices heard because they are in front of the girls.
According to the article above, it seems that you can tell if someone is diabetic or not by their voice 86% of the time. (I’d be interested to know what kind of voice or frequency difference you’re talking about)
It’s amazing how much you can tell from a voice. Like the Beatles’ music separation technology in my last blog, truly, sound technology is becoming amazing.
You may be able to identify various diseases by your voice.
We may even be able to tell if you are in love by your voice.
If we can tell the state of health by voice, perhaps the time will come when voice changers will be used in summit meetings. Or they will only talk via text.
This is because the health of the heads of state is a top secret.（I am reminded of the old story about how the North Koreans even prepared a toilet exclusively for Kim Jong-un during the U.S.-North Korea summit. This was to ensure that health information obtained from defecation would not be leaked to other countries.）
Thus, if we imagine an era in which we can find out a lot of information from voices, we can think of many situations.（ It is also easy to imagine that large companies must be collecting information from many voices around the world via Siri, Alexa, and other smart speakers!）
The voice is not only a unique instrument, but it is also a person’s unique information (health information).