Dear Freesounders,

It looks like 2025 is already over, and many things have happened this year. We celebrated the 20th anniversary of Freesound with a composition contest, a sound installation, and the Freesound Day, which featured talks from people from the Freesound community (the video recordings of the Freesound Day talks are already available!). In the midst of our celebration hangover, it is now time to compute some statistics to summarise 2025 in numbers. The post starts with some numbers and figures following the same post structure that we use every year, and ends with a section in which we analyse the types of sounds uploaded to Freesound using our new Broad Sound Taxonomy (which was introduced in 2025). Without further ado, the number of new sounds uploaded during 2025 has been of…

47,068 new sounds!

which corresponds to…

1494 hours of audio!

In terms of number of sounds, this is 9k less sounds compared to 2024, but in terms of duration, this is ~340 more hours. In fact, the average sound duration is significantly higher in 2025: 114 seconds (compared to 73 in 2024 and 101 in 2023). As usual, these statistics can vary from one year to the next and it does not necessarily mean that there’s a relevant pattern. The influence of individual users is easily significant on a site the size of Freesound.

As a new addition to the “Freesound in numbers” usual stats, here you can see the number of sounds uploaded per month (and the average as a red line). It looks like northern hemisphere spring months are those with more upload activity.

Let’s continue looking at the licenses of uploaded sounds. Here is the Creative Commons license distribution of the newly uploaded sounds:

It is interesting to see how Creative Commons 0 (CC0) peaks at 72%, and Attribution NonCommercial reaches a historic minimum. Last year we observed significant decrease in CC0 licenses sounds, and we hypothesised that could be due to concerns about generative AI model training. Nevertheless, the tendency does not seem to be continuing this year, on the contrary, there have been more CC0-licenses sounds than ever.

With the new additions from 2025, Freesound now currently hosts a stunning total of 714,671 sounds (YES, we surpassed the 700k mark!). The total audio duration is of 536 days and 8 hours. Here is the evolution of the total number of sounds since the beginning of Freesound, and the prediction for the future:

Maybe we can get to 1M by the 25th anniversary of Freesound in 2030? Hmmmm, we’ll probably be quite close anyway!

Here is a tag cloud of the tags of the sounds uploaded during 2025:

Just like every year, we can see some of the classic tags again in the cloud. It is hard to draw strong conclusions just by looking at the picture, but if you look at it closely maybe you’ll find curious tags and some inspiration. The characterisation of the types of sounds uploaded deserves further research so we can have better guesses about any possible existing patterns. Maybe using the Broad Sound Taxonomy could be a good idea for this right? Well, at the end of the post you might find some stats about that 🙂

And now the moment many of you are waiting for: the chart of most prolific sound contributors! You know what? Let’s first pause for a second and listen to some of the best-rated/most-downloaded sounds of 2025:

After this nice intermission, let’s delay it no further: here is the chart of users who have contributed the most sounds in 2025:

	Username	# uploaded sounds		Username	uploaded time (hours)
#1	CVLTIV8R	5943	#1	sensingtheforest	261
#2	looplicator	3214	#2	Sadiquecat	187
#3	sensingtheforest	3135	#3	EtherAudio	98
#4	designerschoice	1917	#4	klankbeeld	73
#5	Sadiquecat	1850	#5	looplicator	53
#6	klankbeeld	900	#6	SieuAmThanh	51
#7	josefpres	731	#7	Philip_Goddard	44
#8	yondercomputer	624	#8	kevp888	27
#9	JW_Audio	598	#9	tim.kahn	22
#10	Cat-Fox_Alex	576	#10	designerschoice	17
#11	kontraamusic	574	#11	BobVoldar	16
#12	Logicogonist	336	#12	CVLTIV8R	15
#13	bassimat	325	#13	josefpres	14
#14	AudioPapkin	269	#14	Matio888	11
#15	DiscordantScraps	222	#15	treytatum3	11
#16	Artninja	222	#16	JW_Audio	11
#17	Dreadwolf910	221	#17	Tonnex	11
#18	kevp888	220	#18	klankschap	11
#19	SieuAmThanh	202	#19	JanRou	11
#20	qubodup	195	#20	logancircle2	11

Let us take this opportunity to thank all sound contributors (not only those appearing in the table)! It is incredible to see year after year that many new sounds are uploaded, and it is also incredible to see how some contributors are so dedicated and make such great efforts to upload very high quality sounds.

But what about downloads? The number of sound downloads (including packs) during 2025 was…

23,729,327 downloads!

This is ~5M more than last year, and ~2M more than two years ago. If we break this into monthly downloads, we get the following:

Similarly to the uploads figure, northern hemisphere spring months are quite active. But there are other spikes, particularly in January (the biggest month in terms of number of downloads). It will be interesting to see why is that, if this is a recurrent pattern, and what type of sounds are downloaded in January. Maybe an idea for next year’s Freesound in numbers post series? All in all, users have downloaded more than 278M sounds and packs from Freesound!

Let’s continue with the term cloud, which shows the most common query terms that have been used when searching in Freesound during 2025:

I’ll let you dig into the term cloud to reach your conclusions but, spoilers alert, it is rather similar to previous years. Last year, for the 2024 in numbers blog post, we included a special section with more details about user queries. We recommend you to check it out if you did not do it (thanks Benno for writing that part last year!)

Now some extra general statistics: In 2025, 15.6k messages were exchanged between 2.4k unique users, 1.1k forum posts were written by 196 users, 485k sound ratings were made by 143k users, and 26.5k sound comments were written by 11k users. These numbers are quite similar to those of last year, which means that the tendency of lowering the number of sound comments but increasing the number of sound ratings is consolidated. We hypothesised in the past that this could happen due to the release of the new UI at the end of 2023, which encourages adding more sound ratings and seems to make the comments section a bit les visible. After the tendency since the release of the new UI, it looks like our hypothesis is correct.

We’re now getting closer to the end of the post. But before finishing, and as it was promised, here is an extra section to deepen our understanding about what types of sounds are being shared in Freesound. And what’s the best way to do that? By classifying them :). Luckily, in April 2025, we introduced the Broad Sound Taxonomy (BST) which provides a structured way to categorize sounds across the platform. The taxonomy is designed to be simple and easy to use, and consists of 5 top-level categories and 23 second-level categories. Since April 2025, when uploading sounds, uploaders must select one of these categories and subcategories. Sounds that were uploaded before the taxonomy was introduced were automatically assigned a category by an algorithm. Here is what the taxonomy looks like:

Using this taxonomy, we analysed Freesound’s uploaded content to classify sounds and identify patterns. A key finding is that uploaded sounds belong to diverse categories, emphasising the variety and heterogeneity of material contributed in Freesound. The figures below illustrate the types of sounds that were uploaded both during 2025 (figure A), and also for all time uploads (figure B). Note that automatic categorization is only used when categories have not been manually provided by sound authors.

Comparing the two figures at the top level of BST (that is to say, considering only the 5 broad sound categories), we can see that both are quite similar. Sound effects are by far the most popular top‑level category in the taxonomy, making up the largest share of uploaded sounds. This dominance likely reflects the broad applicability of isolated sound events (e.g. footsteps, objects, engines) across many creative and practical use cases such as game audio, film post‑production, interactive media, and sound design. Some types of sounds, like music or animal recordings, might be also uploaded on other platforms, but everyday sound effects clearly remain Freesound’s focus. Among the other top-level categories, Music (excerpts, loops, melodies), Instrument samples (single notes, scales), and Soundscapes all show relatively similarly-sized distributions. The balanced mix of them shows that user interests are also heterogeneous. Speech is the least common top‑level category, which raises interesting questions about potential gaps or opportunities on Freesound. Could there be more incentives or challenges to encourage uploads in underrepresented areas like speech recordings?

Zooming in on subcategories (i.e. the second level of BST), certain types of sounds consistently stand out. Objects / House appliances are the most popular, likely because they’re easy to record in everyday life, making them a frequent upload choice for contributors. Solo instrument music (e.g. melodies, passages) is also popular, indicating that users are more likely to upload stems or isolated parts of music rather than full compositions, probably because Freesound isn’t meant for complete music pieces. This year though, there was an increase in Multiple Instrument uploads. Electronic / Design and Experimental sound effects also have a notable share, suggesting uploaders often explore synthesised or processed audio beyond natural recordings. This year we also notice an increase of Nature soundscapes, which could be likely due to individual users rather than a particular generalised new interest on it (probably due to sensingtheforest project uploads?). Lastly, Percussion are the most popular Instrument samples both in 2025 and all-time, but interestingly, there seems to be a decreased interest in uploading traditional instruments such as Winds or Piano in 2025.

There’s more a lot more to discover by looking at how users navigate the taxonomy (remember that in the search page, you’ll find filters for the BST categories!). When searching, each top-level category has an Other subcategory which is used in approximately 7% of sound uploads, and captures uploads where contributors aren’t quite sure which category to choose. That means that we might need to provide some more help to users when choosing their categories. All in all, it’s clear that Freesound is incredibly heterogeneous, and there are tons of sonic gold to discover across every type of sound. There’s a lot more to learn about the distribution of sounds in Freesound, but for now: happy uploading, and don’t forget to give a little thought to where your sounds belong and discover all corners of Freesound 🙂

Aaaaaaand that is all for this year’s post, thanks for reading and we hope you enjoy a 2026 full of sounds!


frederic and penny,
on behalf of the Freesound Team