audio-metadata is a library for reading and, in the future, writing audio metadata.

Clean and understandable code, nice API, and good UX (user experience) are the focal points of audio-metadata. One or more of these things I feel are lacking from already existing alternatives enough to want to write something from scratch with them in mind. Also, there are certain features not present in other solutions that would be prohibitively painful to add.

Features and functionality that set it apart:

• Uses the Python standard load(s)/dump(s) API.
  • Can load filepaths, os.PathLike objects, file-like objects, and bytes-like objects.
• Metadata objects look like a dict and act like a dict.
  • Some common libraries shadow the representation of a dict and/or dict methods but do not behave like a dict.
  • Supports attribute-style access that can be mixed with dict key-subscription.
• All metadata objects have user-friendly representations.
  • This includes humanized representations of certain values like filesize, bitrate, duration, and sample rate.

>>> import audio_metadata

>>> metadata = audio_metadata.load('05 - Heart of Hearts.flac')

>>> metadata
<FLAC ({
	'filepath': '05 - Heart of Hearts.flac',
	'filesize': '44.23 MiB',
	'pictures': [],
	'seektable': <FLACSeekTable (37 seekpoints)>,
	'streaminfo': <FLACStreamInfo ({
		'bit_depth': 16,
		'bitrate': '1022 Kbps',
		'channels': 2,
		'duration': '06:03',
		'md5': '3ae700893d099a5d281a5d8db7847671',
		'sample_rate': '44.1 KHz',
	})>,
	'tags': <VorbisComment ({
		'album': ['Myth Takes'],
		'artist': ['!!!'],
		'bpm': ['119'],
		'date': ['2007'],
		'genre': ['Dance Punk'],
		'title': ['Heart of Hearts'],
		'tracknumber': ['05'],
	})>,
})>

>>> metadata['streaminfo']
<FLACStreamInfo ({
	'bit_depth': 16,
	'bitrate': '1022 Kbps',
	'channels': 2,
	'duration': '06:03',
	'md5': '3ae700893d099a5d281a5d8db7847671',
	'sample_rate': '44.1 KHz',
})>

>>> metadata.streaminfo.bitrate
1022134.0362995076

>>> metadata.streaminfo['duration']
362.9066666666667

>>> metadata['streaminfo'].sample_rate
44100

pip install -U audio-metadata

For the release version, see the stable docs.
For the development version, see the latest docs.

The high-level API and basic usage are covered, but more advanced features/functionality need documentation.

If you're willing to contribute your time to work on audio-metadata, you can:

• Post in the Development category on the Discourse forum.
• Browse and comment on issues or pull requests.
• Open an issue with a bug report or feature request.
• See current projects.
• Contact me by email at mail@thebigmunch.me.

Showing appreciation is always welcome.

Get your own thanks inbox at SayThanks.io.

Contribute by submitting bug reports, feature requests, or code.

Discourse forum