Research Internship, Deezer
Version Identification
Office located in 24 Rue de Calais, Paris
Audio streaming platform
Job no longer available
Audio streaming platform
1001+ employees
Job no longer available
Office located in 24 Rue de Calais, Paris
1001+ employees
Company mission
Connecting fans with artists and creating ways for people to live the music.
Role
Who you are
- Master / PhD student with a background in Computer Science / Computational Linguistics / Applied Mathematics / Statistics
- Strong knowledge of music analysis, natural language processing, applied machine learning and data mining
- Good programming skills for data processing and experimentation (preferred python)
- Creativity and autonomy
What the job involves
- Audio content identification is an essential task for music streaming platforms
- It is used for many purposes, including catalog cleaning, innovative product features, or efficient copyright management
- Witnessing music delivery rates constantly increasing in the last decade, Deezer has been actively involved in developing and improving its identification methods over the years
- In this study, we will focus on musical versions, defined as “any rendition or performance of an existing musical work”
- Identifying two recordings as being versions of the same musical work is quite complex, as many musical dimensions may be heavily affected between both
- For instance, new renditions might alter the harmonic, melodic, rhythmic, timbral, or structural content of the original performance, among others
- This internship will aim at leveraging lyrics at scale in order to identify versions at scale in a large music catalog
Our take
Deezer is a global music-streaming platform built to help people connect with the music they love. Listeners can explore millions of tracks, discover new artists, and enjoy features like Flow, curated playlists, and interactive tools such as Music Quiz and Shaker.
The company operates in a crowded market dominated by giants like SPotify and Apple Music, but it's leaning into personalisation to stand out. Its latest update lets users customise themes, colors, icons and layouts, and gives instant access to their favourite artists, playlists and podcasts through the new Quick Access feature. So, with this approach, it wants to make the app feel more personal in the hopes of appealing to listeners looking for something a bit more tailored.
AI-generated music is another major factor shaping Deezer's direction (and it's a complicated one). Around 50,000 fully AI tracks are uploaded daily, and while many users may be curious, most want them clearly labelled and worry about their impact on real artists. Deezer already detects and labels AI content and keeps it out of recommendations and fraudulent payouts. If it can continue to handle this transparently, AI could become a trust-building advantage rather than a threat for the company.
Freddie
Company Specialist at Welcome to the Jungle
Insights
11% employee growth in 12 months
Company
Funding (last 2 of 7 rounds)
Aug 2018
$185.5m
LATE VC
Jan 2016
$116m
SERIES E
Total funding: $549m
Company HQ
Saint-Georges, Paris, France
Leadership
Alexis Lanternier
(CEO)Previously co-founded Essor where they served as the President for nearly 4 years. Prior to that, they were an EVP of eCommerce at Walmart Canada.
Salary benchmarks
We don't have enough data yet to provide salary benchmarks for this role.
Submit your salary to help other candidates with crowdsourced salary estimates.
Share this job
View 19 more jobs at Deezer