SampleMatch

Using AI to quickly find the perfect drum samples to match your music

SampleMatch - AI that helps you find the perfect drum samples

SampleMatch is an AI model for music creators who are looking for drum samples to match their music. SampleMatch automates the process of finding the most suitable drum samples for any uploaded music track. It uses leading-edge deep learning technology to analyze and sort through large sample libraries to quickly retrieve the most appropriate drum samples for the music. SampleMatch is developed by Music Team at Sony Computer Science Laboratories - Paris (Sony CSL - Paris).

Try demo web app now

SampleMatch overview

Screenshot from the SampleMatch application showing different types of drum samples the user can choose from

SampleMatch is based on an AI model trained using contrastive learning on a data set created from thousands of pop, rock, and electronic music samples.

When an audio track is uploaded to the web application, it is analysed by SampleMatch against the selected drum sample library. In seconds, the drum sample library is sorted into a list of drum kit sounds which, according to the AI model, would best match the uploaded track.

The drum samples in the library are categorized into seven types of drum and cymbal sound, and SampleMatch autopopulates a drum kit, selecting from each of the seven categories.

The samples can be played against the uploaded track and the entire drum kit can be replaced, or different individual samples can be selected manually. The chosen drum samples can then be downloaded.

Why use SampleMatch

State-of-the-art algorithms

Large data set and algorithms developed by Sony.

Find suitable drum samples quickly

Streamline digital music production.

Easy-to-use web app

Effortlessly find the most suitable drum samples based on your uploaded tracks.

SampleMatch demo web application

Try out the SampleMatch demo web application. Upload an audio file and try out the suggested sounds in real time. It takes just a few clicks to preview and audition drum samples that best fit your music.

Watch a short video of the SampleMatch web application, then try the demo web app yourself:

1. Upload the song

Upload your audio file at any stage of the production process.

2. Analyze the song and find the best matches

The deep learning model analyzes the music and shows the drum kit samples that best match the music in this context.

3. Select the samples

Play through the samples and select what to add to a virtual drum kit.

4. Download the drum kit

When you have decided which drum kit samples are best, download and add them to your song.

Who will use SampleMatch

Screenshot from a music editing application
Audio marketplaces
  • Digital Audio Workstation (DAW) software companies
  • Sample library suppliers
Musicians working in a studio
Sound engineers
  • Musicians, DJs, Producers
  • Film and game producers
Musician with headphones playing a musical keyboard and recording it on his laptop
Audio solution providers
  • Audio software developers/Product owners
  • Music and technology researchers
Sony CSL researcher Stefan Lattner is one of the creators of SampleMatch.

The research behind SampleMatch

One of the research team members behind SampleMatch is Stefan Lattner. Stefan is from Sony Computer Science Laboratories - Paris (Sony CSL - Paris) - a research laboratory working on artificial intelligence applied to such areas as language, music, creativity, and sustainability.

Stefan's work often focuses on using computer models to simulate the cognitive processes involved when listening to music, to create better tools for music production and music analysis. He uses machine learning and artificial intelligence with the aim of streamlining the music production workflow, and as a result, creates innovative ways to engage with music and music creation.

For more information about Sony CSL - Paris, please Contact Us.