Hugging Face just launched AI WebTV, an experimental pipeline demonstrating continuous automated video and audio broadcasting. The setup uses ChatGPT to script clip prompts, Zeroscope V2 for video synthesis, FILM for smoothing frames, and MusicGen for background audio, streaming it all out via FFmpeg. Why it matters: It proves that fully autonomous generative broadcasting can run entirely on open-source pipelines.
Epidemic Sound just launched Soundmatch, an AI tool that cuts video soundtracking down to seconds. Editors drop in a video clip and place a marker, and the tool analyzes the visual content and taps into 2.5 billion daily YouTube views worth of data to serve up matching music. It turns hours of manual audio browsing into a one-click recommendation.
The Recording Academy just clarified that only human creators are eligible to win a Grammy Award. Songs with AI vocals or instrumentation can still be nominated, but real people must meaningfully contribute—and songwriting categories must be mostly written by humans. In short, AI can be a tool in the studio, but algorithms won't be walking away with trophies.
Paul McCartney just revealed that AI helped bring John Lennon's voice back for what he calls the final Beatles record. Using audio separation tech from Peter Jackson's documentary, engineers extracted Lennon's vocal cleanly from an old cassette demo. Why it matters: AI is turning damaged, unusable archives into release-ready masters.
Meta just unveiled MusicGen, a generative AI model that turns text prompts and melody inputs into original songs. Trained on 20,000 hours of music and instrument samples, it can theoretically expand a simple hum into a full track. It opens music creation to anyone, though copyright fights with the record industry are likely looming.
A new open-source AI music generator just landed on GitHub, turning simple text prompts and lyrics into complete vocal tracks. Early tests show clean stereo sound across styles like reggae and country, though pronunciation occasionally trips up. The model is now available to download and run locally.
Spotify just yanked tens of thousands of tracks generated by AI startup Boomy, wiping out about 7% of the platform's catalog. The takedown came after Universal Music flagged suspicious bot activity aimed at inflating play counts and siphoning streaming royalties. Spotify confirmed it took action to protect honest artists from artificial streaming scams.
Grimes just launched Elf.Tech, a new AI voice software that lets creators transform vocal recordings into her signature sound. Anyone can use the generated vocals to produce and distribute new tracks, provided they agree to a 50% split on master recording royalties. Why it matters: Rather than policing AI copies, she is turning voice duplication into an open-source revenue share.
Grimes just invited creators to clone her voice with AI, promising a 50% royalty split on any hit track. Free from record label contracts, she announced she wants to "open source all art" and is happy to be a guinea pig, taking a completely different stance from industry crackdowns.
An AI-generated song mimicking Drake and The Weeknd just got wiped from Spotify, Apple Music, and YouTube following takedown demands from Universal Music Group. The viral track racked up over 625,000 plays on Spotify and 275,000 on YouTube before disappearing. It puts the music industry at an urgent crossroads over synthetic voices and copyright law.
Adobe just announced that its Firefly generative AI is expanding to Creative Cloud video and audio apps. While it cannot generate videos from scratch yet, users will soon be able to color grade clips with simple text prompts, generate background audio, and automatically turn scripts into storyboards.
Voicemod just raised $14.5 million to expand beyond gaming soundboards into real-time generative AI voice synthesis. Fresh off integrating tech from its Voctro Labs acquisition, the startup is developing tools that let users sing in someone else's voice or swap vocal identities on calls. Why it matters: Much like viral camera filters changed video, real-time AI avatars are turning personal voice identity into something anyone can customize on the fly.
Google researchers just unveiled MusicLM, a model capable of generating high-fidelity music from complex text prompts and humming. Trained on 280,000 hours of audio, it can weave narrative soundtracks spanning several minutes. But with 1% of outputs directly mimicking copyrighted training data, Google is keeping it locked up for now.
Diffusion models just took over AI image generation, but their reach goes way beyond art. Inspired by thermodynamics, the technology reconstructs data by removing noise step-by-step, guided by OpenAI's CLIP. Why it matters: Researchers are already using diffusion to compose music, design custom proteins, and synthesize DNA sequences.
