From the Blog
Loss Functions in Audio ML
With modern specialized computing power, neural networks that generate audio are
more commonplace. Training these with backpropagation requires a
loss function that can take two audio representations – a model’s current best guess and the true
target sound – and compute a similarity score with differentiable functions.
3 Ways to Classify Drum Sounds
Drums libraries have been one of the most important developments in digital music production. We’re no longer restricted to the stock sounds that come with drum machines, allowing for more room to develop a unique sound. We’ve seen a surge in free drum kits on the web, and there’s an ever-growing list of digital audio workstations with which to use them.
Sometimes vinyl enthusiasts, audio engineers, and music archivists work with audio
files of full album recordings. While it’s possible to split these into individual
track files manually using DAWs and editing software like Audacity,
it’s often a time-consuming process. Introducing Album Split, freeware available for macOS on the app store.
Step 1) Upload or drag and drop your album file.
.mp3 .mp4 .wav .m4a .aac .ac3 .aiff .caf .snd .au are all supported.
Step 2) Album Split analyzes the file and intelligently looks for moments of silence
that can indicate track boundaries. Preview these boundaries to make sure they are
correct, and label each track.
Step 3) Click export, and Album Split will write new files using the detected track boundaries.
Feedback, questions, and support requests can be sent to firstname.lastname@example.org