This is a little hack we did at the HAMR@ISMIR 2015 hackday in Málaga!
- Anna Aljanaki, Stefan Balke, Ryan Groves, Eugene Krofto, Eric Nichols
- Collect several symbolic song datasets, with melody and possibly chords.
- Represent data in a common vector format appropriate for input to a neural net.
- Develop an LSTM architecture for generation ration melody/chord output.
- Make music!
pip install -r requirements.txt
The datasets we use are included. In the following we would like to give credits to the authors:
- Temperley's Rock Corpus
- Essen folk song collection (ESAC): http://www.esac-data.org
- WeimarJazzDatabase: http://jazzomat.hfm-weimar.de
Further documentation can be found on: http://labrosa.ee.columbia.edu/hamr_ismir2015/proceedings/doku.php?id=deepcomposer