The artificial intelligence It never ceases to amaze and, if you thought you had already seen them all with image and video generators, now AudioGen has arrived, an AI that creates sounds from textual commands.
DALL-E 2 and Midjourney have shown how impressive they are at creating art from text, while Meta and Google have debuted their own AI products that generate video. Now, researchers from Meta and the Hebrew University of Jerusalem introduce AudioGen, an AI that generates sounds.
LOOK: Google launches its own AI to create videos from text descriptions
“Whistling with background wind” either “A man speaks while birds sing and dogs bark” are some of the orders that AudioGen follows to create very realistic sounds, as presented by Felix Kreuk, from the research team, through his Twitter account.
We present “AudioGen: Textually Guided Audio Generation”!
AudioGen is an autoregressive transformer LM that synthesizes general audio conditioned on text (Text-to-Audio).
???? Paper: https://t.co/XKctRaShN1
???? Samples: https://t.co/e7vWmOUfva
???? Code & models – soon!(1/n) pic.twitter.com/UiJaA627bv
— Felix Kreuk (@FelixKreuk) September 30, 2022
In their academic paper, the team explains that AudioGen is an autoregressive model of text-based audio generation.
LOOK: The incredible images created by an AI with the phrase: “Miguel Grau in the Huáscar during the War of the Pacific”
According to the researchers, AudioGen can distinguish between different types of noise and separate them from each other; for example, you can filter two people talking at the same time. In this way, the generated samples can be editable and more precise.
The project used 10 data sets so the AI can learn about different soundsyes Although still in development, the team plans release AudioGen to the general publicfor which they will share the code on GitHub.
LOOK: What did Steve Jobs do to not have a license plate on his car without breaking the law?
Image and video generating AIs have already carved out a space for themselves among users, who experiment with these tools on a daily basis with incredible results. Diario El Comercio has even tested what artificial intelligences like Midjourney and DALL-E 2 are capable of creating a futuristic version of the city of Lima.
Source: Elcomercio
I have worked as a journalist for over 10 years and have written for various news outlets. I currently work as an author at 24 News Recorder, mostly covering entertainment news. I have a keen interest in the industry and enjoy writing about the latest news and gossip. I am also a member of the National Association of Journalists.