Careful post-processing of speech and vocal recordings can be a time-consuming effort. The removal of unwanted background noise (which is often present in a recording) is a particularly challenging task. There is a clear need for tools that are easy to use and produce quick results. Accentize (a German developer) approaches the task with a combination of artificial intelligence and signal processing. VoiceGate 2, one of its flagship plug-ins, is getting a major update. VoiceGate2 improves every detail of its predecessor’s functionality.
VoiceGate2 uses machine learning to separate voice from background noise in real time. The underlying artificial neural network is trained on the characteristics of the human voice using thousands of hours of audio data. It’s able to operate with great precision using just a few parameters. The plug-in does everything else independently.
Neural networks tend to require a lot of computing power, but Accentize took care to optimize the plug-in for maximum efficiency. The plug-in supports sample rates up to 192 kHz and runs in DAWs and video editing software (such as Adobe Premiere and Apple Final Cut.)
Compared to the first version, almost every aspect has been completely revised. There’s a new algorithm in place, capable of even better results. The spectral focus mode is also improved. The interface has a fresh new look with additional visualization options. All in all, this is a solid update to a piece of useful and thoroughly developed software.
Accentize VoiceGate2 – Price and availability
Owners of the first version get the update for free. Otherwise you have to pay EUR 154. You can also get the Post Production Bundle (VoiceGate2, DeRoom Pro, Chameleon, and SpectralBalance) for EUR 547. Accentize VoiceGate2 works in 64-bit VST3, AAX and AU formats under Windows and macOS. A demo version and PDF instructions are available from the website.