agangzz / melodyextraction_mcdnn Goto Github PK
View Code? Open in Web Editor NEWThis project forked from keums/melodyextraction_mcdnn
ISMIR2016: Melody extraction on vocal segments using multi-column deep neural networks
This project forked from keums/melodyextraction_mcdnn
ISMIR2016: Melody extraction on vocal segments using multi-column deep neural networks
README.txt ============================================================ ** Contact Info ============================================================ Sangeun Kum <[email protected]> Changheun Oh <[email protected]> Juhan Nam <[email protected]> Korea Advanced Institute of Science and Technology ============================================================ ** Description ============================================================ This is our submission to the 2016 MIREX melody extraction task. The algorithm is a classification based approach using deep neural networks. The file 'main.py' is the main function for calling the algorithm. It takes as parameter, input the full path string for the input file and output file. If you want to know about this algorithms, please check https://wp.nyu.edu/ismir2016/wp-content/uploads/sites/2294/2016/07/119_Paper.pdf ============================================================ ** Platform and Requirements ============================================================ 1. OS : LINUX 2. Programming language : Python 2.7 3. Python Library : 1) Keras (Deep Learning library for Theano) >> http://keras.io/ 2) Theano (Backend of Keras) >> http://deeplearning.net/software/theano/install.html#install 3) Librosa (for audio analysis such as laod,STFT,resampling) >> http://librosa.github.io/librosa/ 4) ffmpeg >> https://www.ffmpeg.org/ >> for install : brew install ffmpeg 5) Numpy, SciPy 4. Hardware 1) GPU : GeForce GTX 980 >> https://developer.nvidia.com/cuda-toolkit 5. Expected runtime : 2~3 seconds/song ============================================================ ** Use ============================================================ The algorithm is called as follows: (to call from the command line) >>python main.py <parameter> <input path> <ouput path> ex) >>python main.py 0.2 '/home/keums/Melody/dataset/adc2004_full_set/file/pop3.wav' './SAVE_RESULTS/pop3.txt' or (to call from the shell) >>main(param = 0.2, PATH_LOAD_FILE='/home/keums/Melody/dataset/adc2004_full_set/file/pop4.wav', PATH_SAVE_FILE='./SAVE_RESULTS/pop4.txt') ** default param = 0.2, if the voice recall rate is low, increaing the param would be effective (0 <= param <= 1 )
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.