1oggenc(1) Vorbis Tools oggenc(1)
2
3
4
6 oggenc - encode audio into the Ogg Vorbis format
7
8
10 oggenc [ -hrQ ] [ -B raw input sample size ] [ -C raw input number of
11 channels ] [ -R raw input samplerate ] [ -b nominal bitrate ] [ -m min‐
12 imum bitrate ] [ -M maximum bitrate ] [ -q quality ] [ --resample fre‐
13 quency ] [ --downmix ] [ --scale ] [ -s serial ] [ -o output_file ] [
14 -n pattern ] [ -c extra_comment ] [ -a artist ] [ -t title ] [ -l album
15 ] [ -G genre ] [ -L lyrics file ] [ -Y language-string ] input_files
16 ...
17
18
20 oggenc reads audio data in either raw, Wave, or AIFF format and encodes
21 it into an Ogg Vorbis stream. oggenc may also read audio data from
22 FLAC and Ogg FLAC files depending upon compile-time options. If the
23 input file "-" is specified, audio data is read from stdin and the Vor‐
24 bis stream is written to stdout unless the -o option is used to redi‐
25 rect the output. By default, disk files are output to Ogg Vorbis files
26 of the same name, with the extension changed to ".ogg" or ".oga". This
27 naming convention can be overridden by the -o option (in the case of
28 one file) or the -n option (in the case of several files). Finally, if
29 none of these are available, the output filename will be the input
30 filename with the extension (that part after the final dot) replaced
31 with ogg, so file.wav will become file.ogg.
32 Optionally, lyrics may be embedded in the Ogg file, if Kate support was
33 compiled in.
34 Note that some old players mail fail to play streams with more than a
35 single Vorbis stream (the so called "Vorbis I" simple profile).
36
37
39 -h, --help
40 Show command help.
41
42 -V, --version
43 Show the version number.
44
45 -r, --raw
46 Assume input data is raw little-endian audio data with no header
47 information. If other options are not specified, defaults to
48 44.1kHz stereo 16 bit. See next three options for how to change
49 this.
50
51 -B n, --raw-bits=n
52 Sets raw mode input sample size in bits. Default is 16.
53
54 -C n, --raw-chan=n
55 Sets raw mode input number of channels. Default is 2.
56
57 -R n, --raw-rate=n
58 Sets raw mode input samplerate. Default is 44100.
59
60 --raw-endianness n
61 Sets raw mode endianness to big endian (1) or little endian (0).
62 Default is little endian.
63
64 --utf8
65 Informs oggenc that the Vorbis Comments are already encoded as
66 UTF-8. Useful in situations where the shell is using some other
67 encoding.
68
69 -k, --skeleton
70 Add a Skeleton bitstream. Important if the output Ogg is
71 intended to carry multiplexed or chained streams. Output file
72 uses .oga as file extension.
73
74 --ignorelength
75 Support for Wave files over 4 GB and stdin data streams.
76
77 -Q, --quiet
78 Quiet mode. No messages are displayed.
79
80 -b n, --bitrate=n
81 Sets target bitrate to n (in kb/s). The encoder will attempt to
82 encode at approximately this bitrate. By default, this remains a
83 VBR encoding. See the --managed option to force a managed
84 bitrate encoding at the selected bitrate.
85
86 -m n, --min-bitrate=n
87 Sets minimum bitrate to n (in kb/s). Enables bitrate management
88 mode (see --managed).
89
90 -M n, --max-bitrate=n
91 Sets maximum bitrate to n (in kb/s). Enables bitrate management
92 mode (see --managed).
93
94 --managed
95 Set bitrate management mode. This turns off the normal VBR
96 encoding, but allows hard or soft bitrate constraints to be
97 enforced by the encoder. This mode is much slower, and may also
98 be lower quality. It is primarily useful for creating files for
99 streaming.
100
101 -q n, --quality=n
102 Sets encoding quality to n, between -1 (very low) and 10 (very
103 high). This is the default mode of operation, with a default
104 quality level of 3. Fractional quality levels such as 2.5 are
105 permitted. Using this option allows the encoder to select an
106 appropriate bitrate based on your desired quality level.
107
108 --resample n
109 Resample input to the given sample rate (in Hz) before encoding.
110 Primarily useful for downsampling for lower-bitrate encoding.
111
112 --downmix
113 Downmix input from stereo to mono (has no effect on non-stereo
114 streams). Useful for lower-bitrate encoding.
115
116 --scale
117 Input scaling factor (helps with clipping inputs).
118
119 --advanced-encode-option optionname=value
120 Sets an advanced option. See the Advanced Options section for
121 details.
122
123 -s, --serial
124 Forces a specific serial number in the output stream. This is
125 primarily useful for testing.
126
127 --discard-comments
128 Prevents comments in FLAC and Ogg FLAC files from being copied
129 to the output Ogg Vorbis file.
130
131 -o output_file, --output=output_file
132 Write the Ogg Vorbis stream to output_file (only valid if a sin‐
133 gle input file is specified).
134
135
136 -n pattern, --names=pattern
137 Produce filenames as this string, with %g, %a, %l, %n, %t, %d
138 replaced by genre, artist, album, track number, title, and date,
139 respectively (see below for specifying these). Also, %% gives a
140 literal %.
141
142 -X, --name-remove=s
143 Remove the specified characters from parameters to the -n format
144 string. This is useful to ensure legal filenames are generated.
145
146 -P, --name-replace=s
147 Replace characters removed by --name-remove with the characters
148 specified. If this string is shorter than the --name-remove
149 list, or is not specified, the extra characters are just
150 removed. The default settings for this option, and the -X option
151 above, are platform specific (and chosen to ensure legal file‐
152 names are generated for each platform).
153
154
155 -c comment, --comment comment
156 Add the string comment as an extra comment. This may be used
157 multiple times, and all instances will be added to each of the
158 input files specified. The argument should be in the form
159 "tag=value".
160
161
162 -a artist, --artist artist
163 Set the artist comment field in the comments to artist.
164
165
166 -G genre, --genre genre
167 Set the genre comment field in the comments to genre.
168
169
170 -d date, --date date
171 Sets the date comment field to the given value. This should be
172 the date of recording.
173
174
175 -N n, --tracknum n
176 Sets the track number comment field to the given value.
177
178
179 -t title, --title title
180 Set the track title comment field to title.
181
182
183 -l album, --album album
184 Set the album comment field to album.
185
186
187 -L filename, --lyrics filename
188 Loads lyrics from filename and encodes them into a Kate stream
189 multiplexed with the Vorbis stream. Lyrics may be in LRC or SRT
190 format, and should be encoded in UTF-8 or plain ASCII. Other
191 encodings may be converted using tools such as iconv or recode.
192 Alternatively, the same system as for comments will be used for
193 conversion between encodings. So called "enhanced LRC" files
194 are supported, and a simple karaoke style change will be saved
195 with the lyrics. For more complex karaoke setups, kateenc(1)
196 should be used instead. When embedding lyrics, the default out‐
197 put file extention is ".oga". Note that adding lyrics to a
198 stream will automatically enable Skeleton (see the -k option for
199 more information about Skeleton).
200
201
202 -Y language-string, --lyrics-language language-string
203 Sets the language for the corresponding lyrics file to language-
204 string. This should be an ISO 639-1 language code (eg, "en"),
205 or a RFC 3066 language tag (eg, "en_US"), not a free form lan‐
206 guage name. Players will typically recognize this standard tag
207 and display the language name in your own language. Note that
208 the maximum length of this tag is 15 characters.
209
210 Note that the -a, -t, -l, -L, and -Y options can be given multiple
211 times. They will be applied, one to each file, in the order given. If
212 there are fewer album, title, or artist comments given than there are
213 input files, oggenc will reuse the final one for the remaining files,
214 and issue a warning in the case of repeated titles.
215
216
218 Oggenc allows you to set a number of advanced encoder options using the
219 --advanced-encode-option option. These are intended for very advanced
220 users only, and should be approached with caution. They may signifi‐
221 cantly degrade audio quality if misused. Not all these options are cur‐
222 rently documented.
223
224
225 lowpass_frequency=N
226 Set the lowpass frequency to N kHz.
227
228
229 impulse_noisetune=N
230 Set a noise floor bias N (range from -15. to 0.) for impulse
231 blocks. A negative bias instructs the encoder to pay special
232 attention to the crispness of transients in the encoded audio.
233 The tradeoff for better transient response is a higher bitrate.
234
235
236 bitrate_hard_max=N
237 Set the allowed bitrate maximum for the encoded file to N kilo‐
238 bits per second. This bitrate may be exceeded only when there
239 is spare bits in the bit reservoir; if the bit reservoir is
240 exhausted, frames will be held under this value. This setting
241 must be used with --managed to have any effect.
242
243
244 bitrate_hard_min=N
245 Set the allowed bitrate minimum for the encoded file to N kilo‐
246 bits per second. This bitrate may be underrun only when the bit
247 reservoir is not full; if the bit reservoir is full, frames will
248 be held over this value; if it impossible to add bits construc‐
249 tively, the frame will be padded with zeroes. This setting must
250 be used with --managed to have any effect.
251
252
253 bit_reservoir_bits=N
254 Set the total size of the bit reservoir to N bits; the default
255 size of the reservoir is equal to the nominal number of bits
256 coded in one second (eg, a nominal 128kbps file will have a bit
257 reservoir of 128000 bits by default). This option must be used
258 with --managed to have any effect and affects only minimum and
259 maximum bitrate management. Average bitrate encoding with no
260 hard bitrate boundaries does not use a bit reservoir.
261
262
263 bit_reservoir_bias=N
264 Set the behavior bias of the bit reservoir (range: 0. to 1.).
265 When set closer to 0, the bitrate manager attempts to hoard bits
266 for future use in sudden bitrate increases (biasing toward bet‐
267 ter transient reproduction). When set closer to 1, the bitrate
268 manager neglects transients in favor using bits for homogenous
269 passages. In the middle, the manager uses a balanced approach.
270 The default setting is .2, thus biasing slightly toward tran‐
271 sient reproduction.
272
273
274 bitrate_average=N
275 Set the average bitrate for the file to N kilobits per second.
276 When used without hard minimum or maximum limits, this option
277 selects reservoirless Average Bit Rate encoding, where the
278 encoder attempts to perfectly track a desired bitrate, but
279 imposes no strict momentary fluctuation limits. When used along
280 with a minimum or maximum limit, the average bitrate still sets
281 the average overall bitrate of the file, but will work within
282 the bounds set by the bit reservoir. When the min, max and
283 average bitrates are identical, oggenc produces Constant Bit
284 Rate Vorbis data.
285
286
287 bitrate_average_damping=N
288 Set the reaction time for the average bitrate tracker to N sec‐
289 onds. This number represents the fastest reaction the bitrate
290 tracker is allowed to make to hold the bitrate to the selected
291 average. The faster the reaction time, the less momentary fluc‐
292 tuation in the bitrate but (generally) the lower quality the
293 audio output. The slower the reaction time, the larger the ABR
294 fluctuations, but (generally) the better the audio. When used
295 along with min or max bitrate limits, this option directly
296 affects how deep and how quickly the encoder will dip into its
297 bit reservoir; the higher the number, the more demand on the bit
298 reservoir.
299
300 The setting must be greater than zero and the useful range is
301 approximately .05 to 10. The default is .75 seconds.
302
303
304 disable_coupling
305 Disable use of channel coupling for multichannel encoding. At
306 present, the encoder will normally use channel coupling to fur‐
307 ther increase compression with stereo and 5.1 inputs. This
308 option forces the encoder to encode each channel fully indepen‐
309 dently using neither lossy nor lossless coupling.
310
311
313 Simplest version. Produces output as somefile.ogg:
314 oggenc somefile.wav
315
316 Specifying an output filename:
317 oggenc somefile.wav -o out.ogg
318
319 Specifying a high-quality encoding averaging 256 kbps (but still VBR):
320 oggenc infile.wav -b 256 -o out.ogg
321
322 Specifying a maximum and average bitrate, and enforcing these:
323 oggenc infile.wav --managed -b 128 -M 160 -o out.ogg
324
325 Specifying quality rather than bitrate (to a very high quality mode):
326 oggenc infile.wav -q 6 -o out.ogg
327
328 Downsampling and downmixing to 11 kHz mono before encoding:
329 oggenc --resample 11025 --downmix infile.wav -q 1 -o out.ogg
330
331 Adding some info about the track:
332 oggenc somefile.wav -t "The track title" -a "artist who per‐
333 formed this" -l "name of album" -c "OTHERFIELD=contents of some
334 other field not explicitly supported"
335
336 Adding embedded lyrics:
337 oggenc somefile.wav --lyrics lyrics.lrc --lyrics-language en -o
338 out.oga
339
340 This encodes the three files, each with the same artist/album tag, but
341 with different title tags on each one. The string given as an argument
342 to -n is used to generate filenames, as shown in the section above.
343 This example gives filenames like "The Tea Party - Touch.ogg":
344 oggenc -b 192 -a "The Tea Party" -l "Triptych" -t "Touch"
345 track01.wav -t "Underground" track02.wav -t "Great Big Lie"
346 track03.wav -n "%a - %t.ogg"
347
348 Encoding from stdin, to stdout (you can also use the various tagging
349 options, like -t, -a, -l, etc.):
350 oggenc -
351
353 Program Author:
354 Michael Smith <msmith@xiph.org>
355
356
357 Manpage Author:
358 Stan Seibert <indigo@aztec.asu.edu>
359
360
362 Reading type 3 Wave files (floating point samples) probably doesn't
363 work other than on Intel (or other 32 bit, little endian machines).
364
365
367 vorbiscomment(1), ogg123(1), oggdec(1), flac(1), speexenc(1), ffm‐
368 peg2theora(1), kateenc(1)
369
370
371
372Xiph.Org Foundation 2008 October 05 oggenc(1)