FuturePulse - Predicting the unpredictable | EN (Breakout Session)
Since we entered the era of big data, the holy grail for all music industry actors have been to find how to use all this data for predicting success. But, this task has proven much more difficult than expected. During three years, a European consortium of companies and universities have worked all around the clock to create new algorithms and models for understanding how we can forecast trends and popularity in music.
Can we predict how many tickets that will be sold for a concert or a festival? Can we predict how a new release will be received in social media and on streaming services? Can we predict genre trends for different markets? Can A&R´s use predictive analytics based on machine learning to detect early trends and get insights for the future?
There are lots and lots of data and statistics out there, millions of graphs. But it is still hard to gain knowledge and really understand what all this data means. FuturePulse has developed a tool that can help in decision making for music industry actors, as well as creating valuable long term know-how, rather than just a lot of data.
During this session you will see how FuturePulse:
Predicts genre popularity trends on different markets based on Spotify Top 200 charts,
How the model Venue Rank predicts which live venues are the best for certain artists when it comes to social media engagement,
How the model T-REC predicts recognition levels of songs, not only popularity, but also how well recognised a song is on a certain market,
How FuturePulse combines open and confidential data (from labels/distributors/radio) into combined timelines, with up to 12 week predictions of what will happen on platforms like Spotify, Apple Music, Deezer, SoundCloud, Facebook, Twitter, YouTube, radio channels globally etc.
Uses advanced Audio Descriptors for music, that identifies such things as moods, situations that are used to predict which songs are the best in playlists,
Automises genre classification based on audio analysis (without the need to tag music manually),
Predicts which playlists that will give the best results for a certain track (based on audio profiling and historic results from similar tracks and artists by maching learning),
Teach the system how certain events influences the specific track/artist (any kind of event, media, tour, release, divorce etc etc), which can then be used to form predictions curves for the same kind of events in the future.
Presented by FuturePulse and We Are Europe.
The Breakout Sessions will not be streamed and are only accessible by registration (the number of seats is limited). To register, you will need the free c/o pop xoxo Convention Online Ticket. After registering for the Online Ticket, you will receive access to our Delegates Database, where you can register for the Breakout Sessions and network and exchange information with other participants.
Click here for the Online Ticket: http://bit.ly/xoxoonlineticket
Do 22.10 16:30
Saki Markovic is working for the indie record label Playground Music Scandinavia where she is in charge of social media and digital marketing across the Nordics. Her task is to create strategies and marketing campaigns to enhance artist’s presence on social media and digital platforms both locally and internationally. She also works with creating and implementing marketing campaigns for Playground’s distributed international labels on the Nordic market. Prior to Playground Music Saki was a part of the digital marketing team at Universal Music Sweden for four years. Within FuturePulse she is leading the Record Label use case providing the project with input and expertise from a label perspective.
Gonçal Calvo is Head of Innovation at the music technology company BMAT, where he has worked for the last 9 years. Graduated as a Telecommunication Engineer from the Universitat Politècnica de Catalunya, he holds a Master's Degree in Acoustics, Signal Processing and Computer Science applied to Music from the IRCAM Institute in Paris. At BMAT, he has coordinated the team in charge of developing the databases and music content management processes, with millions of works, audios and metadata that feed the company's services. He has been responsible for the technical integration of the hundreds of sources that connect to BMAT, including the majors (Sony, Warner, Universal) and the main digital distributors. More recently, as Head of Innovation, Gonçal coordinates BMAT's R+D+i projects in which technologies such as High-Performance Computing, Blockchain or Artificial Intelligence are investigated to improve the company's products and services. When he gets some time, he takes his beloved cello out of the case to refresh a Bach's Suite.
Daniel Johansson is a researcher and analyst at the Linnaeus University in Sweden as well as at the Spotify-backed company Soundtrack Your Brand, which he represents in FuturePulse. His research is focused on the music industrial economy and organisation as well as business economics within the music industry, with a strong focus on streaming and digital distribution. He has a tech background, as a research engineer in computer science at the Royal Institute of Technology in Stockholm, Sweden, as well as working at tech companies in the music industry. He does analyses and writes articles for the music industry news agency www.musikindustrin.se, and does consultancy work for different actors in the music industry. He also plays the guitar like a wiz and occassionaly writes a song or two.