For the first time, robots create sounds that have never appeared before

After incredible achievements such as being able to communicate with each other, making their own companions . the robot can now produce sounds that never existed in the world.

Researchers are extremely talented when creating artificial forms of intelligence that can do almost anything. Robots can beat us in interactive games (board games), drive cars and detect cancer .

Now, artificial intelligence is also capable of creating completely new sounds that no one has ever heard before. These sounds are made possible by advanced mathematical operations combined with samples of real musical instruments.

A long time ago, sometimes you could hear some new sounds coming from the radio. Like researchers say, they are hoping to give musicians a series of new, unlimited melodies, emitted by computer-generated instruments. Having these tools, the composer's work will be much more convenient and wonderful.

This new system, called NSynth , was developed by a team of engineers called Google Magenta . This is a small part of boosting Google's research on artificial intelligence.

"Learning directly from data, NSynth provides artists with intuitive control of timbre and dynamics. At the same time, it also facilitates musicians to discover new and unique sounds. This is extremely difficult or will never work if only hand-controlled synthesizers are used , "the team said.

Robots can create completely new sounds.(Photo: Unsplash).

By accessing the Wired website, users can use some samples made by Nsynth. The first users to use the early music samples were very excited and satisfied.

NSynth samples from about 1,000 different musical instruments and mixes them, but in a very sophisticated and highly complex way.

First, the artificial intelligence program learns to identify the sound characteristics of each instrument, then NSynth proceeds to mix, to create a mix of musical instruments that are not the same. any mix of instruments before.

Sound properties are tuned to produce a kind of sound that is emitted from a completely new device, rather than a mixed sound.

So instead of playing the lute and violin to mix the melody together, you just need Nsynth. It acts as a completely new digital tool based on algorithms to combine the tone of a violin and a flute.

Normally, to create a perfect mix, how many flutes are required by musicians and how many violins? Probably a lot.

Like many other Google artificial intelligence initiatives, NSynth relies on ' deep learing'. This is a specific and detailed approach to artificial intelligence.

Here, they can handle a large amount of data similar to the human brain. That's why these systems are often described as artificial neural networks.

Therefore, not only 'deep learning' systems can use millions of cat photos to accurately identify a cat (an example of the learning ability of artificial intelligence), they can also learn Ask from your mistakes and make progress over time. Artificial intelligence has the ability to train itself to improve skills - just like a human brain.

The idea of ​​deep learning has been around for decades, but now we have only found one type of software and calculations, possibly turning this into reality. One expected result of this is that the 'performances' of NSynth - built by the Google Magenta team.

Nsynth can operate in real time and it also allows creation of new compositions.

Music critic Marc Weidenbaum said in Wired: "Google's new approach to traditional tricks is very clever. Combining the devices shows the promise of music.

"Artically, it can bring some interesting things, and because it's Google's invention, people will follow their lead , " he said.

Google engineers have just introduced NSynth at the Moogfest festival, and people can read the article about their work at arXiv.