New answers tagged machine-learning
3
The NSynth dataset contains labelled audio samples from various instruments over a range of pitches. The data is prepared for use with TensorFlow and also available in JSON/WAV (metadata/audio) format. From the dataset description:
NSynth is an audio dataset containing 305,979 musical notes, each with
a unique pitch, timbre, and envelope. For 1,006 ...
2
A good place to start is browsing the UC Irvine Machine Learning Repository. Note that you can sort by dataset task and attribute types.
Top 50 recent answers are included
