Skip to main content

Use Vamp plugins for audio feature analysis.

Project description

This module allows Python code to load and use native-code Vamp plugins (http://vamp-plugins.org) for audio feature analysis.

The module consists of a native-code extension (“vampyhost”) that provides a low-level wrapper for the Vamp plugin SDK, along with a Python wrapper (“vamp”) that provides a higher-level abstraction.

No code for loading audio files is included; you’ll need to use some other module for that. This code expects to receive decoded audio data of one or more channels, either as a series of frames or as a single whole buffer.

Written by Chris Cannam and George Fazekas at the Centre for Digital Music, Queen Mary University of London. Copyright 2008-2015 Queen Mary, University of London. Refer to COPYING.rst for licence details.

See home page at https://code.soundsoftware.ac.uk/projects/vampy-host for more details.

A simple example

Using librosa (http://bmcfee.github.io/librosa/) to read an audio file, and the NNLS Chroma Vamp plugin (https://code.soundsoftware.ac.uk/projects/nnls-chroma/) for analysis:

$ python
>>> import vamp
>>> import librosa
>>> data, rate = librosa.load("example.wav")
>>> chroma = vamp.collect(data, rate, "nnls-chroma:nnls-chroma")
>>> chroma
{'matrix': ( 0.092879819, array([[  61.0532608 ,   60.27478409,   59.3938446 , ...,  182.13394165,
          42.40084457,  116.55457306],
       [  68.8901825 ,   63.98115921,   60.77633667, ...,  245.88218689,
          68.51251984,  164.70120239],
       [  58.59794617,   50.3429184 ,   45.44804764, ...,  258.02362061,
          83.95749664,  179.91200256],
       ...,
       [   0.        ,    0.        ,    0.        , ...,    0.        ,
           0.        ,    0.        ],
       [   0.        ,    0.        ,    0.        , ...,    0.        ,
           0.        ,    0.        ],
       [   0.        ,    0.        ,    0.        , ...,    0.        ,
           0.        ,    0.        ]], dtype=float32))}
>>> stepsize, chromadata = chroma["matrix"]
>>> import matplotlib.pyplot as plt
>>> plt.imshow(chromadata)
<matplotlib.image.AxesImage object at 0x7fe9e0043fd0>
>>> plt.show()

And a pitch-chroma plot appears.

High-level interface (vamp)

This module contains three sorts of function:

  • Lookup functions: list_plugins, get_outputs_of, get_category_of

    These retrieve the installed plugin identifiers and get basic information about each plugin. For more detailed information, load a plugin and inspect it using the low-level interface described below.

  • Process functions: process_audio, process_frames, process_audio_multiple_outputs, process_frames_multiple_outputs

    These accept audio input, and produce output in the form of a list of feature sets structured similarly to those in the C++ Vamp plugin SDK. The plugin to be used is specified by its identifier. A dictionary of plugin parameter settings may optionally be supplied.

    The _audio versions take a single (presumably long) array of audio samples as input, and chop it into frames according to the plugin’s preferred step and block sizes. The _frames versions instead accept an enumerable sequence of audio frame arrays.

  • The process-and-collect function: collect

    This accepts a single array of audio samples as input, and returns an output structure that reflects the underlying structure of the feature output (depending on whether it is a curve, grid, etc). The plugin to be used is specified by its identifier. A dictionary of plugin parameter settings may optionally be supplied.

    The collect function processes the whole input before returning anything; if you need to supply a streamed input, or retrieve results as they are calculated, then you must use one of the process functions (above) or else the low-level interface (below).

Low-level interface (vampyhost)

This extension contains facilities that operate on Vamp plugins in a way analogous to the existing C++ Vamp Host SDK: list_plugins, get_plugin_path, get_category_of, get_library_for, get_outputs_of, load_plugin, and a utility function frame_to_realtime.

Calling load_plugin gets you a vampyhost.Plugin object, which then exposes all of the methods found in the Vamp SDK Plugin class.

(Note that methods wrapped directly from the Vamp SDK are named using camelCase, so as to match the names found in the C++ SDK. Elsewhere this module follows Python PEP8 naming.)

See the individual module and function documentation for further details.

Python Vamp Host Copyright (c) 2008-2015 Queen Mary, University of London

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the names of the Centre for Digital Music and Queen Mary, University of London shall not be used in advertising or otherwise to promote the sale, use or other dealings in this Software without prior written authorization.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vamp-1.0.0.tar.gz (102.7 kB view details)

Uploaded Source

Built Distributions

vamp-1.0.0-cp27-none-win32.whl (103.0 kB view details)

Uploaded CPython 2.7 Windows x86

vamp-1.0.0-cp27-none-macosx_10_10_x86_64.whl (130.5 kB view details)

Uploaded CPython 2.7 macOS 10.10+ x86-64

File details

Details for the file vamp-1.0.0.tar.gz.

File metadata

  • Download URL: vamp-1.0.0.tar.gz
  • Upload date:
  • Size: 102.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for vamp-1.0.0.tar.gz
Algorithm Hash digest
SHA256 916b72519432566079103670d965ab0017e484086cb4e0e085c9fda131374922
MD5 3dc1fe4d6304ef99caa6f967d85736c6
BLAKE2b-256 382dec94cc7ddee5498baeade734a7457121c16528bc3521085f272eee32d3b8

See more details on using hashes here.

File details

Details for the file vamp-1.0.0-cp27-none-win32.whl.

File metadata

File hashes

Hashes for vamp-1.0.0-cp27-none-win32.whl
Algorithm Hash digest
SHA256 a5e07369ae84b5fedea4f74df77a1f889ef41b9f7fa3ab5d2172f4eedd64431d
MD5 030e39ef4a3c41cce3115bbfd0b3d52c
BLAKE2b-256 a4603fec29bbf0d70fd9f9bc7a35cb5c226412e931d9d90e71f0ac4d0527428d

See more details on using hashes here.

File details

Details for the file vamp-1.0.0-cp27-none-macosx_10_10_x86_64.whl.

File metadata

File hashes

Hashes for vamp-1.0.0-cp27-none-macosx_10_10_x86_64.whl
Algorithm Hash digest
SHA256 5c20fe3e280605ac1a87f0c1d3d67fa56311ae43d44832163cd524d44c7c39fd
MD5 65faf6606cde977f2a701be8cac718d1
BLAKE2b-256 89162f585f3641ce627328ad4a02c115e2d42be5814ce88980190c393d99126b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page