Cedar Audio has announced what it calls “a massive leap forward in AI-powered noise reduction” with the introduction of the VoicEX voice extractor. According to the audio restoration and noise suppression specialists, VoicEX incorporates unique algorithmic capabilities beyond standard AI-based noise reduction techniques. The company adds that, when hearing it for the first time, its own audio engineers were shocked at the amount of noise removed and the quality of the retained voices.
Cedar’s Managing Director Gordon Reid says, “A few years ago, we started developing some of the technologies that would lead us to VoicEX. We were sure these would offer a noticeable improvement over our existing noise reduction techniques, but we didn’t know to what degree. I was stunned when I heard the output from the prototype, and subsequent fine-tuning of the algorithm has been directed at making VoicEX achieve the best possible results over the widest range of audio. The results are remarkable, and we think that VoicEX sets a new standard for real-time audio noise reduction.”
Reid adds that the innovation comes through advances in deep neural networks to identify the voices in the signal and create two audio streams — one containing just the voices and another everything else. “This approach makes VoicEX very simple to use. Where you might expect numerous controls, you’ll find just two knobs – one to determine the loudness of the background and the other to control the loudness of the speech. Reduce the former, and the noise goes away without affecting the timbre of the voices. Reduce the latter, and the voices go away to reveal the background. It’s incredibly simple to understand and takes no time to master and use, but the results are truly amazing.”
You can hear an example of VoicEX in action here.