20 years of sound sharing at Freesound, the prominent international project founded at the UPF

The Music and Technology research group at the UPF launched Freesound in 2005 to provide researchers and creators with access to free sounds, which were very difficult to find on the internet at the time. Used by sound professionals and enthusiasts around the world, Freesound has also become a repository of our sonic memory and is currently facing the challenges of adapting to the AI era.
03.06.2026

Imatge inicial - Image from the Sounds of Barcelona project by Freesound to promote sound culture among the public.

In the different sections of this report, which we present below, you can find out what led to the birth of Freesound and its international impact in the fields of research and audiovisual creation, and on society in general. We take stock of its first twenty years, also presented in the form of a timeline, and offer a final section that projects Freesound into the future in the age of AI. You can access the contents of each section by clicking on the following titles:  


 

Freesound, the world’s first platform to make sounds freely available for research and creation

The sound of footsteps, the sound of musical notes on the piano, the sound of a bird chirping… These are just some of the more than 700,000 files that can be found on the Freesound website, the world’s largest digital platform for finding free sounds. Widely used around the world by professionals in music, film and audiovisual creation, and by researchers in sound technologies, Freesound also boasts a strong following among a large number of enthusiasts. Many people have certainly turned to Freesound to prepare a home video, share recordings of everyday life or prepare the sound effects for an amateur or professional show.

The origins of Freesound are probably less well known, dating back to more than twenty years ago. The idea arose in 2005 at the Pompeu Fabra University (UPF) and, more specifically, in the Music and Technology research group (MTG) of the Department of Engineering. With its extensive background and widespread recognition as a specialist in sound processing technologies, this group spearheaded designing a response to a problem shared by researchers in this field around the world. Until the beginning of the 21st century, researchers found it difficult to freely access sounds. These sounds are the raw material for their studies, since they are essential in order to train artificial intelligence (AI) technologies and tools that are then capable of detecting or comparing them, for example. Strict copyright regulations also hindered access to the material necessary for research.

Xavier Serra, co-founder of Freesound, presents the conference celebrating the project’s 20th anniversary (Poblenou campus, 28/10/25). He is joined by Bram de Jong (left) and Frederic Font (right), co-founder and current coordinator of Freesound, respectively.

The founding of Creative Commons, a window of opportunity to create Freesound

However, when the Creative Commons (CC) international organisation was founded in the United States in 2001, it established a system of standardised licences around the world that allowed each creator to decide under what conditions they wanted to share their works with third parties. This does not mean that all sounds are in the public domain, but rather that the creator can choose which CC licence they want to share their work with. One is effectively equivalent to the public domain (CC0), and this is precisely the most frequent category on Freesound, though there are many more that reserve more or fewer rights to the creator depending on the case.

Following the creation of the CC system, the MTG team saw a window of opportunity to encourage people to share sounds free of charge via the internet and thus meet the needs of numerous researchers and creators around the world. The team did this with the Phonos Foundation, which specialises in promoting cultural activities around music technology and has been linked to the UPF since the creation of the MTG in 1994.

The idea progressed and finally took shape months before the MTG hosted one of the most important sound and music technology conferences. In September 2005, the UPF hosted the 2005 International Computer Music Conference (ICMC) which featured a rather suggestive motto: “Free sound”. In this international forum, the MTG wanted to unveil a project that would serve precisely to facilitate free access to sounds for the research community. The preparatory work started before the conference and the Freesound project was launched on 8 April 2005. Months later, the ICMC would be the international showcase.

Distribution of licences for sounds uploaded to Freesound in 2025

The Creative Commons 0 (CC0) licence assumes that the sounds are free of copyright and are in the public domain. The Attribution-NonCommercial (CC BY-NC) implies that sounds can be used for non-commercial purposes and the original creator must be credited. The Attribution allows the use of the sound for any purpose if the original creator is credited.

Before Freesound was founded, the culture of free software and open data already permeated the MTG. Xavier Serra, the founder and current director of this research group and the president of Phonos, trained in the USA at the prestigious Center for Computer Research in Music and Acoustics (CCRMA) at Stanford, where he absorbed the culture of free software. A great music fan (he plays the cello), Serra is currently a professor at the UPF’s Department of Engineering and a renowned point of reference across the globe in sound technologies. Xavier Serra was one of the two co-founders of Freesound, along with researcher Bram de Jong.

Xavier Serra, co-founder of Freesound: “I think it is a successful example of how the university can respond to needs that go beyond the teaching that we normally do”

Freesound’s other co-founder, Bram de Jong, who currently works as a technology consultant, recalls how quickly the Freesound user community expanded. In addition to researchers and artists, all kinds of enthusiasts of sound and audiovisual creation immediately signed up. The coordinator of the project during the first stage (2005-2010), de Jong was the developer of the first Freesound platform, which was much simpler than the current format, before the current freesound.org web domain was created in 2008. The platform has been improved over time, though the dynamics have always remained essentially the same: after uploading a sound, the user adds a title, tags and a description so that it is easier to locate it later.

Freesound co-founder Bram de Jong recalled how the project’s first web domain had a much simpler structure than the current one, during the project’s 20th anniversary conference (Poblenou campus, 28/10/25).

Since its inception, Freesound has also received numerous awards. For example, in 2006, Xavier Serra and Bram de Jong collected the City of Barcelona Award. The following year, the platform won the first of the four Google Research Awards that it has received throughout its history, which have represented a strong financial injection to develop the project and make it grow.

Xavier Serra (at the lectern) and Bram de Jong (on the left) receiving the City of Barcelona Award for the Freesound project at the Saló de Cent of the City Hall on 13 February 2006 (image from Barcelona City Hall).

However, seeing how Freesound has impacted the day-to-day lives and career paths of many people and its rapid growth around the world has been even more rewarding than the awards.

Total number of sounds uploaded to Freesound and prediction for the future

 

How has Freesound influenced users’ lives? Some personal stories

Users were asked some time ago on one of the platform’s forums to express how Freesound had influenced their lives and the responses were striking. A camera operator, for example, explained that his fondness for Freesound, where he used to share recordings of his partner speaking in various contexts, led him to change jobs and work in sound design.

The project teams also recalls how the mother of a child with autism turned to the Freesound community to help her son. Some autistic children perceive all sounds at the same level, and have difficulty prioritising which specific ones to listen to. In an environment like school, which features a mixture of a great number of sounds, autistic children can feel stressed. In order for the school teachers to understand how the child felt, the mother posted on the Freesound forum asking whether someone could artificially create a sound from a school environment.

A few users responded, and the mother was able to use the sound compositions to explain to the school how her son was feeling. The positive result of this experience made it resonate beyond the child’s school.

Another family helped their dog overcome a phobia of thunder using sounds from Freesound. From early morning until evening, they played the sound of thunder in front of their pet, first very softly and then getting louder and louder. The animal got used to the sound and the dog no longer felt overwhelmed when actual thunder started rumbling. The strategy was so successful that the family ended up receiving requests from other Freesound users to design sound sequences that could help their pets overcome an array of phobias.

These and other cases highlight how Freesound can be used for countless purposes that undoubtedly exceed the creators’ initial expectations.

Read more