Download the installers for your operating system here.

PodcastPlugins.mov

Watch the HD version on youtube

PodcastPlugins

Podcast Plugins are easy-to-use plugins for speech enhancement, specifically designed for podcasters who can't or don't want to dig deeper into the craft of sound engineering.

The bundle consists of two plugins:

PodcastPlugins TRACK

This plugin is inserted on the individual audio tracks of all speakers. It wil take care of level, equalisation and dynamic processing.

screenshot of Podcast Plugin Track

PodcastPlugins MASTER

This plugin is inserted on the audio master bus of the DAW. It will shape the sound some more and take care of levels and dynamics.

screenshot of Podcast Plugin Master

Development

The DSP is written in Faust by Klaus Scheuermann.

The plugins are made by ƒalkTX using DPF.

The logo and signet are by Peter Schlossnikel.

Programming guidance and lovely bits of foss code have been provided by Robin Gareus(x42), Bart Brouns(magnetophon), Jakob Dübel(jkdb), Julius Smith, Dario Sanfilippo and the Faust community. Thanks <3

PodcastPlugins are supported by the PrototypeFund.

Building

Build requirements: gcc or clang as compiler, python and gnu make.

On Linux the following extra development libraries are needed:

OpenGL
X11
Xext
Xrandr

Runtime requirements: OpenGL2 capable graphics card (or software rendering via MESA on Linux).

Once requirements are in place, simply run:

git submodule update --init --recursive
make

Installation

Podcast Plugins are cross-platform and will work on the following operating systems:

macOS
windows
linux
other

Supported plugin formats are CLAP, VST, VST3, LV2 and AU.

Download the installers for your operating system here.

The installer will put the plugin files in the appropriate folders on your system. The next time you start your DAW, they will be available.

macOS

The installer will work for intel macs and silicon/arm macs.

As the installer is not signed, you will have to allow the installer to run. Please learn more for your macOS version here.

windows

tba

linux

An example with the LADSPA plugin illustrates how to install any of the other formats: copy to either the system library in $PREFIX/lib/ladspa with $PREFIX being /usr/local or /usr. For a user-only install copy into $HOME/.ladspa/ or if ~/.local is preferred, declare export LADSPA_PATH=~/.local/lib/ladspa in your ~/.profile and restart your user session. This scheme applies equally to clap, lv2, vst and vst3 too.

Want to use a plugin within a particular software or tool? Ardour loads VST3. Audacity VST, LADSPA and LV2 plugins. ffmpeg and sox can use LADSPA.

other OS

tba

Getting Started

After installation, two plugins are available in your DAW:

PocastPlugins TRACK

TRACK is the actual speech-enhancement plugin. It goes on each individual track of each speaker.

For example:

You recorded two speakers with two microphones to two individual tracks: Please load an instance of TRACK on to each track.

screenshot of DAW with TRACK Plugin

The default setting of the TRACK plugin will be fine in most cases. You can close the plugin window and just let it do it's thing.

If you want to tweak the sound further, read advanced settings.

PodcastPlugin MASTER

MASTER is an automatic leveling and mastering plugin which will take care of the final touch of your podcast. It goes on the master channel of your DAW project.

screenshot of DAW with MASTER Plugin

The default setting of the MASTER plugin will be fine in most cases. You can close the plugin window and just let it do it's thing.

If you want to tweak the sound further, read advanced settings.

Advanced Settings

Advanced Settings for PodcastPlugins TRACK

With the two big knobs in the middle, you can tweak the sound of your speaker:

TIMBRE

TIMBRE controls if the speaker should sound warmer or brighter.

The middle position will work for most voices. It's more or less a question of taste. Extreme settings (+/-5) will still sound good.

STYLE

STYLE controls the estethics of the processing. It ranges from natural to radio-ish.

The middle position will work for most voices.

natural features decent dynamic range control while keeping the natural sound of the voice.

radio-ish features heavier dynamic range compression, known from the sound of speakers on comercial FM radio stations.

INPUT GAIN and TARGET LOUDNESS

INPUT GAIN

Here you can control the volume of the incoming audio before it hits the plugin's processing, although the following leveler stage will take care of the correct loudness anyway. It makes sense when you want to take care of the loudness yourself and the leveler therefore is bypassed.

The green and red brackets indicate a good and bad level.

TARGET LOUDNESS

The standard loudness for podcasts is -16dB LUFS. If for some reason you want to work at a lower loudness (or higher, not recommended), you can set it here.

Advanced Settings for PodcastPlugins MASTER

With the two big knobs in the middle, you can tweak the overall sound of your podcast:

TIMBRE

TIMBRE controls if the overall sound should sound warmer or brighter.

The middle position will work fine in most cases. It's more or less a question of taste. Extreme settings (+/-5) are not so much recommended.

STYLE

STYLE controls the estethics of the processing. It ranges from natural to radio-ish.

The middle position will work fine in most cases. Extreme settings (+/-5) are not so much recommended.

natural features decent dynamic range control while keeping the natural sound of your podcast.

radio-ish features heavier dynamic range compression, known from the sound of comercial FM radio stations.

INPUT GAIN and TARGET LOUDNESS

INPUT GAIN

Here you can control the volume of the incoming audio before it hits the plugin's processing, although the following leveler stages will take care of the correct loudness anyway. It makes sense when you want to take care of the loudness yourself and the leveler therefore is bypassed.

TARGET LOUDNESS

The standard loudness for podcasts is -16dB LUFS. If for some reason you want to work at a lower loudness (or higher, not recommended), you can set it here.

Technical Details

DSP (digital signal processing)

The DSP is written in Faust by Klaus Scheuermann.

PodcastPlugins TRACK

image of signal flow diagram

The plugin's processing consists of 5 stages:

Input Gain: nothing fancy here ;)
Spectral Balancer: The audio is split into 20 bands. Each band is normalized against the wideband loudness of the audio resultig the actual momentary frequency spectrum of the incoming signal. The resulting curve is compared to a target frequency curve carefully chosen based on 20 years of vocal mixing experience. 20 bandpass filters are then adjusted, so the spectrum of the audio matches the target curve. Actually there are two curves to interpolate between with the TIMBRE knob: a warmer one and a brighter one. The intensity of the Spectral Balancer can be adjusted with the strength parameter.
Leveler: a nicely tweaked leveling algorithm with an adjustable target loudness (in LUFS).
Multiband Dynamics: classic multiband compressor/expander with two presets to interpolate between: natural & radio-ish.
Brickwall limiter: a lookahead limiter to avoid peaks.

PodcastPlugins MASTER

image of signal flow diagram

The plugin's processing consists of 5 stages:

Leveler 1: a nicely tweaked leveling algorithm with an adjustable target loudness (in LUFS).
Tilt EQ: a soft tilt EQ consisting of two shelves which can be adjusted with the TIMBRE knob.
Multiband Dynamics: classic multiband compressor/expander with two presets to interpolate between: natural & radio-ish.
Leveler 2: a second leveling stage, very slow and subtile.
Brickwall limiter: a lookahead limiter to avoid peaks.

DPF Plugin Framework

PodcastPlugins use DPF as a plugin framework.

GUI

The GUI requires an OpenGL2 capable graphics card (or software rendering e.g. via MESA on Linux).

It uses a lot of UI elements from DPF-Widgets.

A "theme editor" is available when clicking the plugin name, allowing to customise the UI colors and a few sizes. Clicking "Save" will store it locally on disk so next time the UI is open it will use this as the startup theme.

Check artwork/color_templates on this plugin source code for a few other themes.

CLI

If you prefer batch operations, ffmpeg and sox can use the LADSPA plugin. Two very simple usage examples are:

ffmpeg -i in.wav -filter_complex 'ladspa=pp-master-ladspa:plugin=pp_master' out.wav
sox in.wav out.wav ladspa pp-master-ladspa pp_master 1 1

License

PodcastPlugins are licensed under GPLv3+, see LICENSE for more details.

License

wbob/PodcastPlugins

ErrorLooks like something went wrong!

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Languages

Error
Looks like something went wrong!

Packages