MAN page from OpenSuSE ospac-1.0-3.3.x86_64.rpm
OSPAC
Section: User Commands (1)
Updated: May 2 2016
Index NAME
ospac - open source podcast audio chain
SYNOPSIS
ospac [-voice] [-mix]input1.wav input2.wav...--outputoutput.wav
DESCRIPTION
ospacwill take a multi-channel recording of a conversation and master this toa high-quality mix-down with support for intro and outro. It was developeddue to the need of a batch solution for audio podcast creation.
INFORMATION OPTIONS
- --help
- Display help
- --verbosity [n]
- Set verbosity to level[n]
OUTPUT OPTIONS
- --spatial
- Produce a stereo output with inter aural delay.
- --stereo
- Standard mode to produce a stereo output.
- --mono
- Create a mono output.
- --multi
- Create a multichannel file where each first and further channels ineach input phases are mixed to each other
- --set-stereo-level [n]
- Set maximum volume factor for stereo and spatial mode (0.9) -the lower the stronger the effect is
- --set-stereo-spatial [n]
- Set maximum inter-aural delay distance in meter in spatial mode (0.03) -the higher the stronger - the average distance between human ears are0.22 (meter) so you probably should not go above that limit
OUTPUT TARGETS
- --output [file]
- Write final output to[file]in wave format
- --mp3 [file]
- Write final output to[file]in mp3 format using the externallamecommand which has to be in the path
- --ogg [file]
- Write final output to[file]in ogg format using the externaloggenccommand which has to be in the path
- --plot [file]
- Write a wave form image to[file]in netpbm PPM format
OUTPUT META DATA
- --title [text]
- Set title tag to[text]if this exists in output
- --artist [text]
- Set artist tag to[text]if this exists in output
- --album [text]
- Set album tag to[text]if this exists in output
- --comment [text]
- Set comment tag to[text]if this exists in output
- --category [text]
- Set category tag to[text]if this exists in output
- --episode [text]
- Set episode tag to[text]if this exists in output
- --year [text]
- Set year tag to[text]if this exists in output
- --image [file]
- Set image tag to[file]if this exists in output
NEW SEGMENT OPTIONS
- --voice
- Start of voice channel segment (default)
- --mix
- Start pre-mixed channel segment
- --raw
- Start of raw channel segment
SEGMENT TRANSITION OPTIONS
- --fade [s]
- Fading transition over[s]seconds
- --overlap [s]
- Overlapping transition over[s]seconds
- --parallel
- Render the previous and next segment in parallel. By this, you can haveindependent processing for parallel audio channels. For example, for musicor noise channels. Be aware, that filters such as--skipor--noiseor--trimwill most likely generate out-of-sync audio as the cuts are not synchronizedbetween parallel segments.
CROSSTALK FILTER OPTIONS
- --xgate
- Enable robust crosstalk gate: This tries to mute channels that do notcontribute significant input to cancel crosstalk.Each channel is given an activity score computed by its windowedenergy multiple compared to its assumed silence level. The channelwith highest score has 100%, the rest are scaled down linearilywith their score compared to the highest.
- --no-xgate
- Disable crosstalk gate
- --xfilter
- Enable experimental crosstalk filter: This tries to mute channels thatcontain mainly input that was recorded by other channels before, i.e.crosstalk.
- --no-xfilter
- Disable crosstalk filter for current input section
ADAPTIVE SILENCE SKIP OPTIONS
- --skip
- Soft skip silent passages over 0.5s length
- --skip-level [n]
- Fraction of maximum level considered silence (0.01)
- --skip-order [n]
- Order of time of detected silence that is to be skipped.A level of 0 only skips the 0.5s,a level 0.5 skips about the square root of the silence,a level of 1 skips all of the silence (default: 0.75).
- --no-skip
- Do not skip any content
- --noise
- Skip all actual signal to get an impression of the noise on the channel.
- --trim
- Skip silence from the beginning or the end of channels.
LEVELING, EQUALIZER AND NORMALIZATION OPTIONS
- --leveler
- Enable selective leveler: Enable windowed energy based leveler for current input section.Depending on the energy level in a window around the current positionthe filter decideds whether this is an active channel or not. Incase of an active channel it tries to amplify the current positionsuch the a target energy for the window is achieved. If the channelis deemed inactive, the channel is de-amplified. To avoid to fastchanges in amplification levels, the levels are averaged over time.
- --level-mode {single|stereo|multi}
- The leveling can work on single channels, on each two consecutive stereochannels, or on all channels at once. When using the single mode on a stereosignal, it might destroy wanted differences in loudness between the two channels-therefore they should be joined in the leveling. The multi mode joins all channelsfor the analysis, and levels all channels by the same amount. The single modeis default for voice segments, the stereo mode is default for mix segments.
- --target [n]
- Set average target L2 energy[n]for leveler. The L2 energy is normed to one entry, so a level of2000-6000 seems suitable. Default: 3000.
- --no-leveler
- Disable selective leveler
- --factor [n]
- Multiply channels by factor[n]with sigmoid limiter (default: 1.25):Amplify the signal by given factor and pass the resulting signalthrough a sigmoid function to prevent overdrive, but eventualdistortion cannot be avoided.
- --no-factor
- Disable channel multiplier
- --eqvoice
- Attenuate frequency bands for improved diction
- --no-eqvoice
- Do not attenuate frequency bands
- --analyze
- Analyze frequency band components of active segments.
- --normalize
- Normalize final mix
- --no-normalize
- Disable final normalization
- --lowpass [frequency] [transition]
- Apply a lowpass filter to the current audio segment up to[frequency]Hertz. The[transition]specifies the quality of the filter in Hertz of transition.
- --highpass [frequency] [transition]
- Apply a highpass filter to the current audio segment from[frequency]Hertz. The[transition]specifies the quality of the filter in Hertz of transition.
- --bandpass [low] [high] [transition]
- Apply a bandpass filter to the current audio segment, starting from[low]Hertz up to[high]Hertz. The[transition]specifies the quality of the filter in Hertz of transition.
INPUT AUDIO FILES
- [wave file]
- Load the file[wave file]and regard each channel as an individual input.
- --left [wave file]
- Load the file[wave file]and only use the left channel if it was in stereo.
- --right [wave file]
- Load the file[wave file]and only use the right channel if it was in stereo.
- --to-mono [wave file]
- Load the file[wave file]and use a mono-mixdown of all channels in that file.
- --ascii [sample rate] [text file]
- Load the ascii file[text file]which is assumed to have a sample rate of[sample rate]Hertz. The values can be integer or float values separated bywhite space. The input is rescaled to [-32000,32000] and commentsstarting with '#' are discarded until the next end of line.
EXAMPLES
Mix 2 mono voice recordings with crosstalk filter, leveling and normalization:
- ospac person1.wav person2.wav --output target.wav
Mix podcast with stereo intro and outro:
- ospac --mix in.wav --overlap 4 \ --voice person1.wav person2.wav --overlap 4 \ --mix out.wav --output target.wav
Again with shortened options:
- ospac -mi in.wav -ov 4 -vo person1.wav person2.wav -ov 4 -mi out.wav -out target.wav
Just run the crosstalk filter and create an un-mixed multi-channel output:
- ospac --multi --raw t1.wav t2.wav t3.wav t4.wav --xfilter --output multi.wav
AUTHOR
Sebastian Ritterbusch <ospac at ritterbusch dot de>
SEE ALSO
sox(1),
lame(1),
oggenc(1),
pnmtopng(1),
ppm(5)
Index
- NAME
- SYNOPSIS
- DESCRIPTION
- INFORMATION OPTIONS
- OUTPUT OPTIONS
- OUTPUT TARGETS
- OUTPUT META DATA
- NEW SEGMENT OPTIONS
- SEGMENT TRANSITION OPTIONS
- CROSSTALK FILTER OPTIONS
- ADAPTIVE SILENCE SKIP OPTIONS
- LEVELING, EQUALIZER AND NORMALIZATION OPTIONS
- INPUT AUDIO FILES
- EXAMPLES
- AUTHOR
- SEE ALSO
This document was created byman2html,using the manual pages.
