wavrw is a command line tool to read (and someday write) WAV files with a focus on visualizing the structure of files and parsing metadata.
It's initially intended as a learning project with aspirations to be generally useful. It's currently in very early days and may not be useful to anyone else.
Current status: Most chunks from the initial WAV spec RIFF1991 are supported. Only supports reading metadata. Doesn't yet support writing metadata. Doesn't yet read or write audio data (samples).
I intend to eventually support every WAV chunk I can find a sample of. If there's something you'd like supported, please let me know by adding an issue. If you have them, please include a link to a specification and one or more example files.
Take care, -Brian
wavrw answers questions about WAV files:
- What is actually in this file?
- Does another program preserve all WAV chunks when saving and exporting files? (by running it before and after the other program and comparing the output)
$ wavrw view test_wavs/example_a.wav`
test_wavs/example_a.wav:
offset id size summary
12 fmt 16 PCM (0x1), 1 chan, 24/48000
36 bext 604 BWDate, BWTime, BWDescription
648 data 1440 audio data
2096 LIST-adtl 70 labl(3)
2174 ID3 2048 ...
4230 SMED 8812 ...
13050 LIST-INFO 214 IPRD, IGNR, ISFT, INAM, IARL, ICOP, IART, ICMT
13272 iXML 4516 ...
17796 cue 76 3 cue points
17880 _PMX 3706 ...
21594 MD5 16 0x37A5BED4393B8F3708963F5E59C7F483
21618 CSET 8 code_page: (0), United States of America(1), En ...
$ wavrw view --format=detailed test_wavs/example_a.wav
test_wavs/example_a.wav:
offset id size summary
12 fmt 16 PCM (0x1), 1 chan, 24/48000
| format_tag : WAVE_FORMAT_PCM (0x1)
| channels : 1
| samples_per_sec : 48000
| avg_bytes_per_sec : 144000
| block_align : 3
| bits_per_sample : 24
--------------------------------------
36 bext 604
| description : BWDescription
| originator : BWOriginator
| originator_reference : BWOriginatorRef
| origination_date : BWDate
| origination_time : BWTime
| time_reference : 0
| version : 1
| umid : 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
| loudness_value : 0
| loudness_range : 0
| max_true_peak_level : 0
| max_momentary_loudness : 0
|max_short_term_loudness : 0
| coding_history : 22
--------------------------------------
648 data 1440 audio data
2096 LIST-adtl 70 labl(3)
| labl : 1, Region 01
| labl : 2, Marker 01
| labl : 3, Marker 02
--------------------------------------
2174 ID3 2048 ...
4230 SMED 8812 ...
13050 LIST-INFO 214 chunk: text
| IPRD : CDTitle
| IGNR : Category
| ISFT : Soundminer
| INAM : TrackTitle
| IARL : BWOriginator
| ICOP : TrackYear Manufacturer (Library) URL
| IART : Artist
| ICMT : Description
--------------------------------------
13272 iXML 4516 ...
17796 cue 76 name: position, chunk_id, chunk_start, block_start, sample_offset
| 1 : 0, data, 0, 0, 0
| 2 : 240, data, 0, 0, 240
| 3 : 360, data, 0, 0, 360
--------------------------------------
17880 _PMX 3706 ...
21594 MD5 16 0x37A5BED4393B8F3708963F5E59C7F483
21618 CSET 8 code_page: (0), United States of America(1), English(9), US(1)
| code_page : 0
| country_code : United States of America(1)
| language : English(9)
| dialect : US(1)
--------------------------------------
$ wavrw help
WAV file metadata read/write utility
Usage: wavrw <COMMAND>
Commands:
view Summarize WAV file structure and metadata
list List directories of files, show single line summary of chunks
topic Print additional help and reference topics
help Print this message or the help of the given subcommand(s)
Global Options:
-h, --help Print help
-V, --version Print version
$ wavrw topic chunks
bext Broadcast Extension for motion picture, radio and television.
CSET Character set information. Code page, language, etc. Very Rare.
cue A series of positions in the waveform data chunk.
data Audio samples. This parser skips all audio data (for now).
fact Number of samples for compressed audio in data.
FLLR Padding, filler or outdated information.
fmt Format of audio samples in data.
inst Pitch, volume, and velocity for playback by sampler.
JUNK Padding, filler or outdated information.
MD5 Checksum of audio data of the WAVE.
PAD Padding, filler or outdated information.
plst Play order for cue points. Very rare.
RIFF Container structure for multimedia data.
smpl Information needed for use as a sampling instrument.
LIST-adtl CuePoint annotation chunks.
file Information embedded in other file formats.
labl A label, or title, to associate with a CuePoint.
ltxt Text associated with a range of data samples.
note Comment text for a CuePoint.
LIST-INFO Descriptive text chunks:
IARL, IGNR, IKEY, ILGT, IMED, INAM, IPLT, IPRD, ISBJ,
ISFT, ISHP, IART, ISRC, ISRF, ITCH, ICMS, ICMT, ICOP,
ICRD, ICRP, IDPI, IENG, ISMP, IDIT.
LIST-wavl Audio and/or silence chunks:
data Audio samples. This parser skips all audio data (for now).
slnt A count of samples to be played as silence.
For more information about chunks, see wavrw's sibling project WavRef:
https://wavref.til.cafe/