{
    "info": {
        "author": "Andy Pearce",
        "author_email": "andy.pearce@surrey.ac.uk",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 2 - Pre-Alpha",
            "License :: OSI Approved :: Apache Software License",
            "Programming Language :: Python :: 2.7",
            "Topic :: Multimedia :: Sound/Audio :: Analysis"
        ],
        "description": "## AudioCommons Timbral Models\nThe timbral models were devleoped by the [Institute of Sound Recording (IoSR)](http://www.iosr.uk/projects/AudioCommons/) at the University of Surrey, and was completed as part of the [AudioCommons project](https://www.audiocommons.org).  \n\nThe current distribution contains python scripts for predicting eight timbral characteristics: *hardness*, *depth*, *brightness*, *roughness*, *warmth*, *sharpness*, *booming*, and *reverberation*.  \n\nMore detailed explanations of how the models function can be found in Deliverable D5.8: Release of timbral characterisation tools for semantically annotating non-musical content, available: http://www.audiocommons.org/materials/\n\n\n## Installing the package\nThe timbral_models package can be installed using the pip command.  This will handle installation of all dependencies.  In the update to version 0.4, the dependency to essentia was removed and only pip installable packages are required.\n```\npip install timbral_models\n```\n\nPlease note that during testing, pip was unable to install some of the dependencies and produced an error.  In these cases, either rerun the `pip install timbral_models` command or install the offending dependency directly, e.g. `pip install numpy`. \n\nThe package can also be installed locally and be made editable.  To do this, clone the repository, navigate to the folder, and run the pip command `pip install -e .`.  In this method, dependencies will not be installed.\n  \n\n## Dependencies\nThe script can also be downloaded manually from the github repository (https://github.com/AudioCommons/timbral_models).  If doing this, dependencies will need to be manually installed.  The timbral models rely on several other easily accessible python packages: `numpy`, `soundfile`, `librosa`, `sklearn`, and `scipy`.  These are all easily installed using the `pip install` command.  e.g.\n```\n$ pip install numpy\n$ pip install soundfile\n$ pip install librosa\n$ pip install scipy\n$ pip install sklearn\n$ pip install six\n$ pip install pyloudnorm   \n```\n\n\n## Using the models\nThe models are formatted in a python package that can be simply imported into a Python script.\nThe timbral extractor can be used to extract all timbral attributes with a single function call.\n\nTo calculate the timbral attributes, pass the timbral extractor function a string of the filename.  The method will then read in the audio file internally and return all timbral characteristics.\n```\nimport timbral_models\nfname = '/Documents/Music/TestAudio.wav'\ntimbre = timbral_models.timbral_extractor(fname)\n```\nIn this example, `timbre` will be a python dictionary containing the predicted *hardness*, *depth*, *brightness*, *roughness*, *warmth*, *sharpness*, *booming*, and *reverberation* of the specified audio file.  \n\n\n### Single attribute calculation\n\nAlternative, each timbral attribute can be calculated individually by calling the specific timbral function, e.g. `timbral_hardness(fname)`.\nThese are named `timbral_xxx(fname)`, where `xxx` represents the timbral model, and also require a string of the filename to be analysed.\n```\nimport timbral_models\nfname = '/Documents/Music/TestAudio.wav'\ntimbre = timbral_models.timbral_hardness(fname)\n```\n\n\n## Model output\nThe *hardness*, *depth*, *brightness*, *roughness*, *warmth*, *sharpness*, and *booming* are regression based models, trained on subjective ratings ranging from 0 to 100.  However, the output may be beyond these ranges.   \nThe `clip_output` optional parameter can be  used to contrain the outputs between 0 and 100.\n```\ntimbre = timbral_models.timbral_extractor(fname, clip_output=True)\n```   \nFor additional optional parameters, please see Deliverable D5.8.\n\nThe *reverb* attribute is a classification model, returning 1 or 0, indicating if the file \"sounds reverberant\" or \"does not sound reverberant\" respectively.\n\n\n## MATLAB Reverb model\nAlso contained in this repository is a full version of the timbral reverb model.  For instruction on installing and using this, please see Deliverable D5.8.\n\n## Version History\nThis section documents the version history of the timbral models.  To download a specific version of the model that relate to a specific deliverable, please check this section and download the most recent version from that date.\n\n2019/01/24 - Version 0.4 of timbral_models, relates to Audio Commons Deliverable D5.8.  This version of the repository relates to the software version 0.4 on PyPI.\n\n2018/12/14 - Version 0.3 of timbral models, relates to Audio Commons Deliverable D5.7. This version of the repository relates to the software version 0.3 on PyPI.\n\n2018/07/26 - Version 0.2 of timbral models, relates to Audio Commons Deliverable D5.6.  This version of the repository relates to the software version 0.2 on PyPI. \n\n2017/09/05 - Version 0.1 of timbral models, relates to Audio Commons Deliverable D5.3.  This version of the repository relates to the software version 0.1 on PyPI.\n\n2017/04/27 - Version 0.0 of the timbral models, relates to Audio Commons Deliverable D5.2. \n\n\n## Citation\nFor refencing these models, please reference Deliverable D5.8, available: http://www.audiocommons.org/materials/",
        "description_content_type": "",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/AudioCommons/timbral_models",
        "keywords": "",
        "license": "Apache Software",
        "maintainer": "",
        "maintainer_email": "",
        "name": "timbral_models",
        "package_url": "https://pypi.org/project/timbral_models/",
        "platform": "",
        "project_url": "https://pypi.org/project/timbral_models/",
        "project_urls": {
            "Homepage": "https://github.com/AudioCommons/timbral_models"
        },
        "release_url": "https://pypi.org/project/timbral_models/0.4.0/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Algorithms for predicting the timbral characteristics of audio files",
        "version": "0.4.0"
    },
    "last_serial": 4740829,
    "releases": {
        "0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "f57d9fa9996afb73a26eeb67f346f5ca",
                    "sha256": "5f7717d31684ed419979c5d4db588bf42e65c6e10ca1b272709dde8d5bf37047"
                },
                "downloads": -1,
                "filename": "timbral_models-0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "f57d9fa9996afb73a26eeb67f346f5ca",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 16313,
                "upload_time": "2017-04-28T09:19:28",
                "url": "https://files.pythonhosted.org/packages/7d/e5/d24929416d4c0f0fb928647727184353f3500e30d654264321ee14c52ef7/timbral_models-0.1.tar.gz"
            }
        ],
        "0.2.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d12538a5dfd16ae89825c94878002a90",
                    "sha256": "97ad954824358eed377c19288521cafb379c1d0802b2c0774cb240c37da80111"
                },
                "downloads": -1,
                "filename": "timbral_models-0.2.0.tar.gz",
                "has_sig": false,
                "md5_digest": "d12538a5dfd16ae89825c94878002a90",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 29369,
                "upload_time": "2018-07-26T13:45:20",
                "url": "https://files.pythonhosted.org/packages/b8/fe/bb39d5d39e7a48cba58c44324b88c8639adb50df54771a98e19dde3eb657/timbral_models-0.2.0.tar.gz"
            }
        ],
        "0.2.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "9a8de68187998e5c24059cbac8a24046",
                    "sha256": "c250870f0df20e27afd33b038f185222ef8949bf236049f8f0bf279ddb853cd0"
                },
                "downloads": -1,
                "filename": "timbral_models-0.2.1.tar.gz",
                "has_sig": false,
                "md5_digest": "9a8de68187998e5c24059cbac8a24046",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 30865,
                "upload_time": "2018-07-26T14:01:29",
                "url": "https://files.pythonhosted.org/packages/d1/c9/1950a4138dd6ba69bf6640aa8f7d098a93965f28af0070474f1ac331bc87/timbral_models-0.2.1.tar.gz"
            }
        ],
        "0.2.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "e3e396e49538a674c100b672bdae00ab",
                    "sha256": "08be491ed4acf98997b1a3b97d8a5cc7b5909d1606f0a15af0e40e3aa56560ca"
                },
                "downloads": -1,
                "filename": "timbral_models-0.2.2.tar.gz",
                "has_sig": false,
                "md5_digest": "e3e396e49538a674c100b672bdae00ab",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 30845,
                "upload_time": "2018-07-26T14:18:23",
                "url": "https://files.pythonhosted.org/packages/71/57/a74ca0519f9445ae14af157eb0fa4585a4c2181a78831d63163ba3a85a20/timbral_models-0.2.2.tar.gz"
            }
        ],
        "0.2.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "11f899d2f96166bce449dc341cb6a102",
                    "sha256": "0790f5152f50569458e4ac6ce82b774e64f10add963338ad6a0638488d231f54"
                },
                "downloads": -1,
                "filename": "timbral_models-0.2.3.tar.gz",
                "has_sig": false,
                "md5_digest": "11f899d2f96166bce449dc341cb6a102",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 31124,
                "upload_time": "2018-07-26T14:32:52",
                "url": "https://files.pythonhosted.org/packages/b2/15/ce80111f719c5e99cbe38f5374975cfd2c185767e34c3892c8c5978a3619/timbral_models-0.2.3.tar.gz"
            }
        ],
        "0.2.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "bc5e14ea823b8f669b8c7c8db29afc7a",
                    "sha256": "a7e9e7f293d7871f7df0886f9af3e9bad5a00054ea2138b63f2fd4bfc9783ebb"
                },
                "downloads": -1,
                "filename": "timbral_models-0.2.4.tar.gz",
                "has_sig": false,
                "md5_digest": "bc5e14ea823b8f669b8c7c8db29afc7a",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 32515,
                "upload_time": "2018-08-16T13:58:38",
                "url": "https://files.pythonhosted.org/packages/93/69/28fb979d515b2c743d309fa3dce9c8c660733809792369b01cd6f55efdbe/timbral_models-0.2.4.tar.gz"
            }
        ],
        "0.3.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "086169bf9e118ccae017deb34c8e9d0c",
                    "sha256": "a5ec8b67938910dd30bc24337d5678634bb80b59f1b3e6d2dba5d4faf36d8424"
                },
                "downloads": -1,
                "filename": "timbral_models-0.3.0.tar.gz",
                "has_sig": false,
                "md5_digest": "086169bf9e118ccae017deb34c8e9d0c",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 34956,
                "upload_time": "2018-12-14T17:59:15",
                "url": "https://files.pythonhosted.org/packages/b4/57/054170ac6085b513fb018bff9a32a9d377a81c92d0421ac67a2b2c72e7fb/timbral_models-0.3.0.tar.gz"
            }
        ],
        "0.4.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d2c3f7ec988615846f77e404be9145c3",
                    "sha256": "07c284e1454ba5f5d249d2997f5fc10a7d73a2710547a3d96540cdbd38cdcb09"
                },
                "downloads": -1,
                "filename": "timbral_models-0.4.0.tar.gz",
                "has_sig": false,
                "md5_digest": "d2c3f7ec988615846f77e404be9145c3",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 59945,
                "upload_time": "2019-01-25T16:49:25",
                "url": "https://files.pythonhosted.org/packages/4c/50/d94415f4145c5c7ef8782d27ab920ca47e6fb703a6c38e45c4b0b512b5b8/timbral_models-0.4.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "d2c3f7ec988615846f77e404be9145c3",
                "sha256": "07c284e1454ba5f5d249d2997f5fc10a7d73a2710547a3d96540cdbd38cdcb09"
            },
            "downloads": -1,
            "filename": "timbral_models-0.4.0.tar.gz",
            "has_sig": false,
            "md5_digest": "d2c3f7ec988615846f77e404be9145c3",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 59945,
            "upload_time": "2019-01-25T16:49:25",
            "url": "https://files.pythonhosted.org/packages/4c/50/d94415f4145c5c7ef8782d27ab920ca47e6fb703a6c38e45c4b0b512b5b8/timbral_models-0.4.0.tar.gz"
        }
    ]
}