InterBEE REVIEW2016
53/71

Sound restoring technology and application1:00 PM - 4:00 PM, Friday, November 18th Audio SymposiumAudio restoration workflowThis was presented by Mr. Jonathan Wyner (Chief Engineer, M Works Mastering Studio). Mr. Wyner explained the basic concepts of audio restoration, and introduced techniques to restore and correct visual expression using the iZotope RX software’s spectrogram. His demonstration included removing steady state noise (hissing and humming noises) and transient noise, as well as audio restoration by correcting the head room of distorted materials.Audio restoration software and mastering in music productionThis was presented by Mr. Masahiro Tabayashi (Chief Mas-tering Engineer, Studio Engineering Department, NIPPON COLUMBIA CO., LTD.). Audio restoration when mastering includes the removal of performance noise, lip noise, hum-ming noise, hissing noise and so on. When reproducing a former sound source (archive material) as a product, noise processing is important, even essential, so audio restoration software matching the noise type is used for processing. A noise processing demonstration was performed at the venue using the sound source of a standard-playing record.Effects of original sound reproduction technology in real-timeThis was presented by Mr. Takahiro Sutoh (Manager, Micro-sound). He introduced the CEDAR Audio product, “dns2”, which muffles noises from outside, such as the sounds of cicadas and rain, as well as background music at events, etc. Microphones were actually taken into the venue as a test sound source, and a pin microphone was used to show how it can be applied to howling prevention. The foremost attraction of dns2 is being able to process virtually in real-time, and it can be fully used at sites where real-time processing is required.Demixing, artificial intelligence and the future of audio restorationThis was presented by Mr. Jason E. Davies (President & CEO, Eleven Dimensions Media, LLC). Zynaptiq develop solu-tions using technology based on source separation, machine learning, pattern recognition, and Mr. Davies introduced their future capabilities.He gave an explanation of UNVEIL (real-time reverb adjust-ment and signal focusing), UNCHIRP (reusing audio with degraded sound quality caused by irreversible voice coding), and UNMIX::DRUMS (emphasizes or removes drum sound in a mixed sound source) and demonstrated their respective capabilities.83

元のページ  ../index.html#53

このブックを見る