annotate gmm_baseline_experiments/external_libs/librosa/librosa-0.3.1/docs/install.rst @ 5:b523456082ca tip

Update path to dataset and reflect modified chunk naming convention.
author peterf
date Mon, 01 Feb 2016 21:35:27 +0000
parents cb535b80218a
children
rev   line source
peterf@2 1 Installation instructions
peterf@2 2 =========================
peterf@2 3
peterf@2 4 The simplest way to install *librosa* is through the Python Package Index (PyPI). This
peterf@2 5 will ensure that all required dependencies are fulfilled. This can be achieved by
peterf@2 6 executing the following command::
peterf@2 7
peterf@2 8 pip install librosa
peterf@2 9
peterf@2 10 or::
peterf@2 11
peterf@2 12 sudo pip install librosa
peterf@2 13
peterf@2 14 to install system-wide.
peterf@2 15
peterf@2 16 If you've downloaded the archive manually from the `releases
peterf@2 17 <https://github.com/bmcfee/librosa/releases/>`_ page, you can install using the
peterf@2 18 `setuptools` script::
peterf@2 19
peterf@2 20 tar xzf librosa-0.3.tar.gz
peterf@2 21 cd librosa-0.3/
peterf@2 22 python setup.py install
peterf@2 23
peterf@2 24 Additional notes
peterf@2 25 ----------------
peterf@2 26
peterf@2 27 By default, *librosa* will use `scipy.signal` to resample audio signals, which can
peterf@2 28 be slow in practice. It is highly recommended to install `librsamplerate
peterf@2 29 <http://www.mega-nerd.com/SRC/>`_ and the corresponding python module,
peterf@2 30 `scikits.samplerate <https://pypi.python.org/pypi/scikits.samplerate>`_.
peterf@2 31
peterf@2 32 Once these are installed, *librosa* will use the faster `scikits.samplerate` for all
peterf@2 33 resampling operations.