Monday, December 11, 2023

 

Software DJ creates automated pop song mashups #Acoustics23


Automated software mixes drums, vocals to create unique musical combinations


Reports and Proceedings

ACOUSTICAL SOCIETY OF AMERICA

Algorithm works to isolate and blend individual components from multiple songs 

IMAGE: 

THE ALGORITHM WORKS TO ISOLATE AND BLEND INDIVIDUAL COMPONENTS FROM MULTIPLE SONGS TO PRODUCE A UNIQUE COMPOSITE WITH A PLEASING SOUND. CREDIT: XINYANG WU

view more 

CREDIT: THE ALGORITHM WORKS TO ISOLATE AND BLEND INDIVIDUAL COMPONENTS FROM MULTIPLE SONGS TO PRODUCE A UNIQUE COMPOSITE WITH A PLEASING SOUND. CREDIT: XINYANG WU




SYDNEY, Dec. 7, 2023 – Song mashups are a staple of many DJs, who mix the vocals and instrumentals from two or more tracks into a seamless blend, creating a new and exciting final product. While the result is fun to listen to, the creation process can often be challenging, requiring knowledge and expertise to select the right tracks and mash them together perfectly.

Xinyang Wu from the Hong Kong University of Science and Technology took a different approach, designing a computer algorithm to intelligently create mashups using the drum tracks from one song and the vocals and instrumentals from another. He will present his work Dec. 7 at 4:20 p.m. Australian Eastern Daylight Time, as part of Acoustics 2023, running Dec. 4-8 at the International Convention Centre Sydney.

While some algorithms and automated software can attempt to create song mashups, their results are often clunky and unrefined. These methods layer the complete, unaltered tracks on top of each other, aligning them based on detected key moments in the music, rather than skillfully combining the vocals and instrumentals of different songs.

“Imagine trying to make a gourmet meal with only a microwave – that’s sort of what automated mashup software is up against compared to a pro chef, or in this case, a professional music composer,” said Wu. “These pros can get their hands on the original ingredients of a song – the separate vocals, drums, and instruments, known as stems – which lets them mix and match with precision.”

His algorithm takes a different approach, mimicking the process used by professionals. The software works to isolate the stems from each song and identify the most dynamic moments. It adjusts the tempo of the instrumental tracks and adds the drum beat mashup at exactly the right moment for maximum effect.

The result is a unique blend of pleasing lyrics and exciting instrumentals with wide-ranging appeal.

“From what I’ve observed, there’s a clear trend in what listeners prefer in mashups,” said Wu. “Hip-hop drumbeats are the crowd favorite – people seem to really enjoy the groove and rhythm that these beats bring to a mashup.”

Now that the software has been tested on drum tracks, the team plans to tackle bass mashups next. For Wu, the dream is to expand the algorithm to incorporate the full instrumental suite and put user-friendly mashup technology directly into the hands of listeners.

“Our ultimate goal is creating an app where users can pick any two songs and choose how to mash them up – whether it’s switching out the drums, bass, instrumentals, or everything together with the other song’s vocals,” said Wu.

###

----------------------- MORE MEETING INFORMATION -----------------------

The Acoustical Society of America is joining the Australian Acoustical Society to co-host Acoustics 2023 in Sydney. This collaborative event will incorporate the Western Pacific Acoustics Conference and the Pacific Rim Underwater Acoustics Conference.

Main meeting website: https://acoustics23sydney.org/  
Technical program: https://eppro01.ativ.me/src/EventPilot/php/express/web/planner.php?id=ASAFALL23    

ASA PRESS ROOM

In the coming weeks, ASA’s Press Room will be updated with newsworthy stories and the press conference schedule at https://acoustics.org/asa-press-room/

LAY LANGUAGE PAPERS

ASA will also share dozens of lay language papers about topics covered at the conference. Lay language papers are summaries (300-500 words) of presentations written by scientists for a general audience. They will be accompanied by photos, audio, and video. Learn more at https://acoustics.org/lay-language-papers/.

PRESS REGISTRATION

ASA will grant free registration to credentialed and professional freelance journalists. If you are a reporter and would like to attend the meeting or virtual press conferences, contact AIP Media Services at media@aip.org. For urgent requests, AIP staff can also help with setting up interviews and obtaining images, sound clips, or background information.

ABOUT THE ACOUSTICAL SOCIETY OF AMERICA

The Acoustical Society of America (ASA) is the premier international scientific society in acoustics devoted to the science and technology of sound. Its 7,000 members worldwide represent a broad spectrum of the study of acoustics. ASA publications include The Journal of the Acoustical Society of America (the world’s leading journal on acoustics), JASA Express Letters, Proceedings of Meetings on Acoustics, Acoustics Today magazine, books, and standards on acoustics. The society also holds two major scientific meetings each year. See https://acousticalsociety.org/.

ABOUT THE AUSTRALIAN ACOUSTICAL SOCIETY

The Australian Acoustical Society (AAS) is the peak technical society for individuals working in acoustics in Australia. The AAS aims to promote and advance the science and practice of acoustics in all its branches to the wider community and provide support to acousticians. Its diverse membership is made up from academia, consultancies, industry, equipment manufacturers and retailers, and all levels of Government. The Society supports research and provides regular forums for those who practice or study acoustics across a wide range of fields The principal activities of the Society are technical meetings held by each State Division, annual conferences which are held by the State Divisions and the ASNZ in rotation, and publication of the journal Acoustics Australia. https://www.acoustics.org.au/ 

###


No comments: