Wavdumper

Author: Kristian Ovaska (kristian.ovaska [at] helsinki.fi)
WWW: http://www.cs.helsinki.fi/u/hkovaska/wavdumper/
License: GNU General Public License
Current version: 0.5 (2003-11-13)

Wavdumper prints detailed information on the headers and structure of a WAV file. It is written in Python and has been tested on Python 2.2, 2.3 and 2.4. In order to understand the output of Wavdumper for complex WAV files, it helps to be familiar with the WAV file format. See the References section for pointers.

Recognized chunk types: cue, data, disp, fact, fmt, junk, labl, ltxt, list, note, pad, plst, smpl

Wavdumper does not read actual sound data; it is for inspecting headers only. For programmers, try the wave module in the standard Python library instead.

Download

zip files have Windows-style line endings, tar.gz files have Unix-style line endings.

Wavdumper 0.5: wavdumper-0.5.zip (15,958 bytes) / wavdumper-0.5.tar.gz (14,095 bytes)

Usage

From command line:

> wavdumper.py file1.wav [file2.wav ...]

or (wildcards work on Windows, too):

> wavdumper.py *.wav

As a Python module:

>>> import wavdumper
>>> wav = wavdumper.Wav('filename.wav')
>>> wav.printInfo()

or

>>> wav.printInfo(outfile) # print to custom file-like object

Release history

Version 0.5 (2003-11-13)

• First public release

Notes

• Recognizes big-endian WAV files (with RIFF tag 'RIFX'), but I've never seen one and couldn't test it.
• There are some checks for WAV file validity (for example, that there is no more than one data chunk), but they aren't complete in the sense that they would catch all invalid files.
• Performance with large files is no problem, since Wavdumper skips through sound data.

Todo

• Add support for more chunks: inst, ...
• Verbose option: print every value in the headers.
• Print durations more nicely: for example, use minutes instead of seconds for long files.
• Some values (like SMPTE stuff) are not printed, they probably should.
• Some text fields in WAV files may contain line breaks, which breaks indentation of Wavdumper output. This looks a bit ugly.
• Make more complete test cases.

Example output

> wavdumper.py test1.wav

File: test1.wav (1910 bytes)
Chunk at pos 20: id = "fmt ", length = 16 bytes
  Format Chunk
  Data format: Uncompressed PCM
  Channels: 1
  Sample rate: 8800 Hz
  Average bytes per sec: 17600
  Block align (bytes): 2
  Bits per sample: 16
Chunk at pos 44: id = "data", length = 1760 bytes
  1760 bytes sample data
  880 samples
  0.100 seconds
Chunk at pos 1812: id = "smpl", length = 60 bytes
  Sampler Chunk
  Manufacturer: 0
  Product: 0
  MIDI note (60=middle-C): 69 (A-5) +- 0.0%
  Sample loop
    Loop ID: 0
    Type: forward
    Range: 0 - 879
    Repeat: infinity
Chunk at pos 1880: id = "LIST", length = 30 bytes
  List Chunk, id = "INFO"
  Chunk at pos 1892: id = "INAM", length = 17 bytes
    Name: 440 Hz sine wave

References

Contain overlapping information.

• http://www.sonicspot.com/guide/wavefiles.html
• http://www-mmsp.ece.mcgill.ca/Documents/AudioFormats/WAVE/Samples.html
• http://www.borg.com/~jglatt/tech/wave.htm
• http://ccrma-www.stanford.edu/CCRMA/Courses/422/projects/WaveFormat/
• RFC 2361 for a list of format codes (many are rare or obsolete)