This paper summarizes some recent advances on a set of tasks related to the processing of singing using state-of-the-art deep learning techniques. We discuss their achievements in terms of accuracy and sound quality, and the current challenges, such as availability of data and computing resources. We also discuss the impact that these advances do and will have on listeners and singers when they are integrated in commercial applications.

This paper from Graves et al is a cool extension of Variational Autoencoders. I found no existing implementation so I wrote one in pytorch.

Scientists at MIT have created an AI psychopath trained on images from a particularly disturbing thread on Reddit. Norman is designed to illustrate that the data used for machine learning can significantly impact its outcome. "Norman suffered from extended exposure to the darkest corners of Reddit, and represents a case study on the dangers of Artificial Intelligence gone wrong when biased data is used in machine learning algorithms," writes the research team. Norman is trained on image captioning, a form of deep learning that lets AI generate text descriptions of an image. Norman learned from image captions of a particularly disturbing subreddit, dedicated to images of gore and death.

Phil Rosenthal, the host of Netflix's "Somebody Feed Phil," meets Jefferson Graham for lunch, and reveals his top 3 favorite destinations in the world. Phil Rosenthal is the star of Netflix's travel documentary series "Somebody Feed Phil" (Photo: Jefferson Graham) LOS ANGELES -- At age 58, Phil Rosenthal finally has a website, https://www.philrosenthalworld.com The star of Netflix's globe-trotting "Somebody Feed Phil," travel documentary series had a previous life as the co-creator of "Everybody Loves Raymond," and was happy to have people look him on his IMDB page. But now that he's got a second act as an internet video star, with "Phil," he realized he needed more, as his fans were demanding it. "They would sit with a pencil, writing the names of the places we visited as the show was on," says Rosenthal.

Westworld just wrapped up its second season on HBO, and even after 20 episodes, fans of the show like science fiction editor John Joseph Adams are still no closer to understanding how the show's guns are able to kill robots but not humans. "The creators must have some idea how these guns work," Adams says in Episode 316 of the Geek's Guide to the Galaxy podcast. I want to know what they think, how they work. Because it doesn't make any sense to me." Season 2 includes a passing reference to "sim bullets," which makes Geek's Guide to the Galaxy host David Barr Kirtley think that there must be something special about the bullets themselves. "Maybe all the bullets have little incendiary things in them that cause them to self-destruct if they're flying in the direction of a guest," he says. "But in one scene they just hold the gun right against someone's chest, and I don't see how the bullet's not going to kill you from that range, even if it is sort of programmed to self-destruct." Writer Sara Lynn Michener wonders if maybe it's the guns that are special rather than the bullets. "You can have a gun that has paintball bullets in it, and you can have a gun that has real bullets in it," she says, "and the gun determines, 'All right, who am I aiming at?', and decides which bullet to release based on that." But science fiction author Anthony Ha says that even if there is an explanation for how the guns work, he still doesn't understand how humans are kept safe from other weapons such as arrows and axes. "It definitely drives me crazy," he says. "Do they have safeties on the swords here too?

How to solve every single Fortnite Season 4 challenge and level up to Tier 100. Fortnite's 4th season is almost over. Season 5 starts on July 12th with a brand new Battle Pass and 100 more tiers to level up, unlocking new skins, emotes, sprays and more along the way. In order to get to tier 100 you'll either need to pay or play your way to the top and unlock the Omega skin. In any case, in order to help you all reach that coveted tier 100 and mop up any missing challenges you might have, we've put together this guide, which includes a solution to every single Season 4 Battle Pass challenge.

Music is there when we wake up in the morning and turn on the radio or TV. Music is in the films we see, online videos we experience, TV shows we watch, games we play, advertisements we see and hear: it brings together life's experiences. It's core to many different types of entertainment; music is in nearly everything. Our research shows that the music creation industry has grown 38% in the past four years, and is on pace to meet or exceed $15 billion in 2018. Music drives film, TV, games, online video, advertising, stage plays, and news services.

Music is there when we wake up in the morning and turn on the radio or TV. Music is in the films we see, online videos we experience, TV shows we watch, games we play, advertisements we see and hear: it brings together life's experiences. It's core to many different types of entertainment; music is in nearly everything. Our research shows that the music creation industry has grown 38% in the past four years, and is on pace to meet or exceed $15 billion in 2018. Music drives film, TV, games, online video, advertising, stage plays, and news services.

This is part one of a three-part tutorial series in which you will use R to perform a variety of analytic tasks on a case study of musical lyrics by the legendary artist, Prince. Musical lyrics may represent an artist's perspective, but popular songs reveal what society wants to hear. Lyric analysis is no easy task. Because it is often structured so differently than prose, it requires caution with assumptions and a uniquely discriminant choice of analytic techniques. Musical lyrics permeate our lives and influence our thoughts with subtle ubiquity. The concept of Predictive Lyrics is beginning to buzz and is more prevalent as a subject of research papers and graduate theses. This case study will just touch on a few pieces of this emerging subject. To celebrate the inspiring and diverse body of work left behind by Prince, you will explore the sometimes obvious, but often hidden, messages in his lyrics. However, you don't have to like Prince's music to appreciate the influence he had on the development of many genres globally. Rolling Stone magazine listed Prince as the 18th best songwriter of all time, just behind the likes of Bob Dylan, John Lennon, Paul Simon, Joni Mitchell and Stevie Wonder. Lyric analysis is slowly finding its way into data science communities as the possibility of predicting "Hit Songs" approaches reality. Prince was a man bursting with music - a wildly prolific songwriter, a virtuoso on guitars, keyboards and drums and a master architect of funk, rock, R&B and pop, even as his music defied genres. In this tutorial, Part One of the series, you'll utilize text mining techniques on a set of lyrics using the tidy text framework.

The Belgian artificial intelligence company ScriptBook is one such entity that could completely alter the industry. ScriptBook's technology works by analyzing the text of screenplays to produce box office forecasting for a film. In a presentation at the Karlovy Vary International Film Festival, the company's founder, Nadira Azermai, outlined how ScriptBook's algorithms could have rejected scripts for Sony movies that were box office failures. Sony released 62 movies from 2015 to 2017, 32 of which lost money. By retroactively analyzing the screenplays, ScriptBook's system correctly identified 22 of the 32 films that ultimately lost money.