Write a Python wrapper for matlab/c code

The attached file '[url removed, login to view]' contains a matlab wrapper to C code. The project is to port the matlab wrapper to a Python wrapper. This is a library for estimating fundamental frequency (pitch) information from sounds. The c code should be arranged into a single shared library, and then wrapped for Python. The shared library should be compileable in windows and linux, and detailed instructions on how to compile on both platforms should be included with the source code. NumPy arrays should be underlying data type. The project should be useable in Python 2.7, and 3.2 at least. I don't need the plotting functions, just the data.

## Deliverables

1) There should be a c shared library that compiles on both linux (so) and windows (.dll), and the project should include clear instructions on how to compile for both, as well as pre-compiled versions. This lib should contain all of the main code for yin. If you are in a position to advise as to the possibility of compiling on Mac, that would be great, although not required.

2) There should be a wrapper written in python to the shared lib. The wrapper should work with both Python 2.7 and 3.2. It should take a numpy array as input, and return a set of numpy arrays that represent the output of the yin algorithm.

3) I don't need the plotting functions that are found in the matlab version, just the data ([url removed, login to view], [url removed, login to view], [url removed, login to view], [url removed, login to view], [url removed, login to view], etc). The output data should match that of the (included) matlab version (i will provide matlab data - input and output - for comparison if needed). I use the [url removed, login to view] datatype, and read and write wavefiles like this (please use these for comatibility):

from [url removed, login to view] import wavfile

import numpy as np

def wavread(filename):

fs,data = [url removed, login to view](filename);

return [url removed, login to view](data/32768.), fs

def wavwrite(data,fs,filename):

[url removed, login to view](filename, fs, [url removed, login to view](data*32768))

4) The only dependencies should be numpy and the standard python library. Docstrings should be included, as well as other standard programming "good" practices, and examples of how to use it.

音频服务 Odd Jobs 脚本安装 shell脚本 软件构架

项目ID： #2722262

关于项目