Meta has now released five new AI models to the public – all dealing with familiar things like image-to-text and image generation. But there’s also a new audio generation model, Jasco, which Meta says “is comparable to the evaluated baselines considering generation quality.”
Source: Meta launches new AI music model; Jen has “ethically-trained” AI model
Given the podcast titles that typically top the charts, you might assume that true crime and chat shows are the most popular podcast genres in the world. However, according to MIDiA’s recent report, “Defining podcast taste: The universal appeal of music, news, and comedy”, these genres are actually some of the least popular globally. There is only one podcast genre for which global audiences indicate universal preference: music.
Source: The verdict is in: Music is podcasting’s biggest missed opportunity
Among AI music generators capable of creating entire songs with just a prompt, Suno and Udio are considered by many to be among the best. They are also considered by many to have used copyrighted music to train their AI models, without authorization. Now, the Recording Industry Association of America (RIAA) is coordinating lawsuits on behalf of major recording companies against the two AI companies.
Source: Major record companies sue AI music generators Suno, Udio for ‘mass infringement’ of copyright
The EU’s Digital Markets Act, often called the DMA, enacted far-reaching restrictions on big technology companies and digital competition. One requirement was for interoperability, meaning developing software that could work across operating systems and hardware, making it easier for consumers to move their data or switch providers.
Source: Apple Says Regulatory Concerns Might Prevent Rollout of AI Features in Europe
TikTok is forming an in-house Music Content Investment Team based in Los Angeles, New York, and San Jose, focusing on “partnership or acquisition opportunities in the music content space on a global level.” In other words, TikTok is moving into the competitive music M&A market. Tellingly, it’s looking to the world of investment banking and private equity to find the execs who will lead this charge.
Source: TikTok is forming an investment team to acquire music content and companies
Queen’s music catalog is among the most valuable of the rock era — with classics like “Bohemian Rhapsody,” “Another One Bites the Dust,” “Radio Ga Ga,” “39,” “Somebody to Love” and “You’re My Best Friend” as well as the perennial stadium-shakers “We Will Rock You” and “We Are the Champions,” the songs are globally popular and enormously lucrative. One other player was said to be very close in the bidding, but stopped short at $900 million.
Source: Queen Catalog to Be Acquired by Sony Music for £1 Billion
Last week, AI startup Luma posted a series of videos created using its new video-generating tool Dream Machine, which the company describes as a “highly scalable and efficient transformer model trained directly on videos.” The only problem? At about 57 seconds in, the Dream Machine-generated trailer for Monster Camp features a slightly AI-smudged but still recognizable Mike Wazowski from Pixar’s Monsters, Inc.
Source: An AI video tool just launched, and it’s already copying Disney’s IP
Startups need large amounts of data, storage and chip capacity to be able to train their AI chatbots, which is where regulators believe the tech giants hold too much power. There is some evidence that smaller players are being forced into exclusive, opaque deals to run technology from Nvidia, Microsoft and their rivals, which can give the already dominant players an even bigger advantage.
Officially launching this summer, MicDrop is a real-time (AU, VST3, AAX) plug-in compatible with all major digital audio workstations (DAWs). UMG and SoundLabs’ collaboration will enable UMG artists to create custom vocal models that will be available for their exclusive creative use cases, and not available to the general public.
Source: SoundLabs and UMG Announce Agreement for AI Voice Modeling