• The Journal of the Acoustical Society of Korea
• /
• v.33 no.1
• /
• pp.60-67
• /
• 2014

The purpose of this study is to suggest a new concept of the absolute sound level system in order to basically solve all the irrational aspects from the mastering stage to the output stage during the process of producing a sound source. Through this absolute sound level system, all the input absolute sound levels of sound sources become standardized based on 60 dB(S), and then sound source producers can implement a sound level they intend by using the differential sound level tag. Besides, by matching the output absolute sound level with the input absolute sound level, a target sound level can be implemented as a listener intends. With this system introduced, sound source producers are able to focus on the completion of a sound source itself without an unnecessary competition for sound levels, and only by inputting the differential reduction tag, they can implement a sound level balance as intended. At the same time, listeners are able to appreciate all kinds of sound sources fit for the standard of absolute s ound levels they want to listen to, without a process of trial and error, no matter what kinds of listening environments and sound systems they have.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.424-434
• /
• 2020

This paper describes an algorithm that calculates Absolute Sound Level (ASL) for contents platform. ASL is a single volume representing individual sound sources and is a concept designed to integrate and utilize the sound level units in digital sound source and physical domain from a speaker in practical areas. For this concept to be used in content platforms and others, it is necessary to automatically derive the ASL without having to go through a hearing of mastering engineers. The key parameters of which a person recognizes the representative sound level of an individual single sound source are the areas of "frequency, maximum energy, energy variation coefficient, and perceived energy distribution," and the ASL was calculated through the normalizing of the weights.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.738-739
• /
• 2023

음원별 음량 차이가 심하여 청각적 불편함을 초래하는 상황을 음량정규화 기술을 통해 완화하려고 한다. 그러나 현재 상용화된 관련 기술을 보면 대부분 음량을 조정하는 경우에 음원의 품질을 보장할 수 없거나, 반대로 음원의 품질을 보장하는 경우에 음량 조정이 지극히 미세한 수준이다. 또한 음원 제공 플랫폼이나 음원의 형태에 따라 적용 가능한 대상이 제한되는 문제도 존재한다. 이러한 품질 문제를 해결하기 위해 본 논문에서는 쾌적최대음량(CML)과 절대음량 체계를 도입하고 음원 손실을 최소화한 음량평준화 시스템을 제안한다. 그리고 일반적인 음원 파일뿐만 아니라 실시간 스트리밍 콘텐츠의 음원을 음량정규화할 수 있도록 설계하여 기존 상용 기술의 대상 제한 문제를 완화하고자 한다. 본 논문은 청자 혹은 서비스 이용자의 만족스러운 음원 감상에 기여한다는 점에서 의의를 가진다.