search

MTG / essentia

C++ library for audio and music analysis, description and synthesis, including Python bindings
http://essentia.upf.edu
GNU Affero General Public License v3.0
2.83k stars 530 forks source link

beat aligned summarization #370

Closed echo66 closed 8 years ago

echo66 commented 8 years ago

Greetings!

Is there any example with beat aligned summarization? I'm following the normal and streaming tutorials in python but, the closest thing seems to be: (1) create a pool aggregator per beat interval. (2) add the feature vectors to the corresponding aggregators. Right now, AFAIK, this is done "manually" (i.e.: there are no Essentia classes to take care of this).

dbogdanov commented 8 years ago

This can be done manually in a few lines of code in python. You create pools for each beat interval and feed them the frames.

In standard mode this is a question of extracting beats from audio and then iterating through audio (FrameGenerator) to compute frames and populate the corresponding pools. You can also use Slicer to cut audio into segments first, and then compute frames on each segment.

echo66 commented 8 years ago

Thanks for the clarification, Dmitry. :+1: