Open Source Podcast Audio Chain
C++ Makefile Python Roff Shell
Switch branches/tags
Nothing to show
Latest commit 76e0035 Nov 25, 2017 Sebastian Ritterbusch Sebastian Ritterbusch Information on building ospac with libav/ffmpeg support.
Permalink
Failed to load latest commit information.
GUI-Release
Release-ffmpeg Support for ffmpeg. Nov 25, 2017
Release Added project wide version setting in file version. Apr 26, 2016
examples Test result updates. Apr 12, 2016
images Skip target example images May 8, 2016
logo Created a multi-resolution icon for ospac-gui. May 3, 2016
src Removed aac parameter from build without aac support. Nov 25, 2017
test Included skip target tests May 8, 2016
.gitignore Improve tab completion May 28, 2016
Doxyfile Activated batchmode to ignore doxygen induced latex errors Mar 13, 2016
LICENSE Initial commit Mar 8, 2016
Ligo.md Added reference Mar 20, 2016
Makefile
README.md Information on building ospac with libav/ffmpeg support. Nov 25, 2017
Stereo.md Added reference to Sendegate and removed wrong option in --multi case. Mar 20, 2016
bash_completion.sh Improve tab completion May 28, 2016
create_tab_completion.py Improve tab completion May 28, 2016
footer.html Initial commit Mar 8, 2016
makefile.targets Added extra linux target "ospac-gui-linux" to overcome static linking… May 2, 2016
ospac-internal.pdf Documentation update. Nov 25, 2017
ospac.1 Added documentation for --skip-target option. May 8, 2016
version Release 1.0.1 Nov 25, 2017
zsh_completion.sh Improve tab completion May 28, 2016

README.md

Synopsis

Ospac will take a multi-channel recording of an audio podcast conversation and master this to a high-quality mix-down with support for intro and outro.

Features

Beside other features, ospac includes a robust leveler, two solutions to avoid crosstalk and soft silence skipping.

Leveler

Ospac provides an energy based leveler. The following plot shows the original on top and the levelled outcome on the bottom:

Example for the leveling algorithm: Original top, leveling bottom

Crosstalk Filter

This waveform shows a crosstalk from the first channel in to the second channel, and some original content in the second half of the second channel:

Example for crosstalk

The included crosstalk gate is a very robust algorithm that results in the following:

Result of crosstalk gate

A new experimental crosstalk filter tries to improve the result:

Result of crosstalk filter

Soft silence skipping

When enabled, ospac may reduce silence above 0.5s to shorter pauses:

Result of soft silence skipping

Alternatively, ospac can try to reach target lengths (90%, 80%, 70%, 60%, and 50%) compared to the original:

Result of target silence skipping

Of course, the result will only be really acceptible until all silence space is used up.

Examples

Mix 2 mono voice recordings with crosstalk filter, leveling and normalization:

ospac person1.wav person2.wav --output target.wav

Mix a 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

Motivation

Ospac was developed due to the need of a batch solution for audio podcast creation. It is a rewrite and compilation of the scripts and methods used for the Modellansatz podcast, http://modellansatz.de/ .

Installation

Check that you have libsndfile and a c++-compiler on your system. Then, you may build and install ospac in this way:

$ make ospac && sudo make install

If you have ffmpeg and libav on your system, you may also build ospac with support for loading of various audio formats and writing aac files:

$ make ospac-ffmpeg && sudo make install-ospac-ffmpeg

On MacOS you can install libsndfile using homebrew simply by:

$ brew install libsndfile

If you want to build the GUI, install the development packages for fltk-1.3 and then build ospac-gui:

$ make ospac-gui && sudo make install-gui

Since there seem to be some broken fltk-1.3 source packages on linux, you can alternatively try the following to build a non-statically linked version:

$ make ospac-gui-linux && sudo make install-gui

To simplify usage tab-completion files for bash and zsh can be generated:

$ make tab-completion

The tab-completion is then usable by appending the following to your .[bash|zsh]rc file:

source PATH_TO_OSPAC/[zsh|bash]_completion.sh

API Reference

See the doxygen docs https://github.com/sritterbusch/ospac/raw/master/ospac-internal.pdf for a detailed API reference.

Tests

You may run tests that create exemplary mixdowns and waveform images that are used in the API documentation.

$ make test

License

The MIT License (MIT)

Copyright (c) 2016-2017 Sebastian Ritterbusch ospac@ritterbusch.de

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.