Adobe just made its Firefly audio suite generally available, letting creators generate commercially safe music, voiceovers, and sound effects directly in their workflow. The update also integrates Google's Gemini Omni Flash model and rolls out a free tier with daily generations for Firefly AI Assistant. Why it matters: Content creators can now bypass licensing nightmares and voice recording logistics inside a single platform.
Stability AI just launched beta tools for Stable Audio 3.0, including a native DAW plugin and an overhauled web studio. The macOS AU/VST3 plugin syncs directly to your session BPM in Logic Pro and Ableton Live to generate up to six minutes of audio, while the web app adds multi-track mixing. Outputs are commercially safe and user-owned.
Alibaba just rolled out HappyShrimp 1.0 into beta, an AI music generation model that crafts complete tracks—melody, arrangement, lyrics, and vocals—from a single text description. It maps out musical structure and emotional tone across genres like pop, rock, and Chinese style, with China's Taihe Music already on board for artist co-creation.
Pika just launched Pika Audio, a family of four foundation models covering video soundtracks, music, sound effects, and text-to-speech. The company claims its training and inference stack makes them up to 20x cheaper than rival models, with speech generation 9x cheaper than ElevenLabs v3. They are available exclusively through the Pika API Club.
Suno just launched Studio 2.0, upgrading its browser-based workstation with MIDI editing, automation curves, and advanced stem separation. Musicians can now use a chat interface to design custom audio plugins from scratch or use MIDI inputs as generation prompts. The update is currently available to Premier subscribers.
MiniMax just launched Music 3.0, an open-weights AI model that generates complete songs up to five minutes long in a single pass. Powered by a hybrid architecture based on Qwen3.5-8B alongside flow matching, it preserves long-range song structure while capturing realistic instrument dynamics. Why it matters: It brings fully produced, extended AI tracks into the open-source fold without sacrificing acoustic fidelity.
Xiaomi researchers just open-sourced MiDashengLM-Gen, an end-to-end framework capable of generating coherent audio scenes blending speech, background music, sound effects, and ambient noise in a single pass. Powered by a fine-tuned Qwen3-1.7B and Flow Matching DiT, it turns structured text prompts into unified 16 kHz audio under an Apache 2.0 license.
Anthropic, Spotify, and Substack are rolling out new transparency tools to tackle the AI authenticity crisis. Anthropic is now watermarking Claude-generated text and C2PA files to comply with the EU AI Act, while Spotify is tagging non-human artists with badges and cutting them from recommendations. Why it matters: With nearly all listeners unable to spot AI music, platforms are racing to restore credibility.
Spotify just introduced new "AI Persona" badges to flag artist profiles where the identity appears to be AI-generated rather than a real human. Launching mid-September, these profiles will be excluded from editorial and algorithmic recommendations by default unless listeners actively follow them. Artists can self-disclose now via Spotify for Artists or face platform reviews.
Suno just published a set of AI music principles and announced new guardrails to fight fraud across the music ecosystem. The company is rolling out tamper-resistant audio watermarking and fingerprinting to help platforms spot AI-generated songs, alongside an upcoming downloads policy aimed at curtailing mass streaming uploads.
Spotify just announced that independent licensing giant Merlin, representing over 30,000 labels, has joined Universal Music Group on its upcoming AI remix and covers tool. The feature lets fans generate AI covers and remixes exclusively from consenting artists, ensuring original creators receive credit and compensation. While AI-generated tracks flood streaming platforms, Spotify is pitching this as the first licensed path for interactive fan music, launching first as a paid add-on in private preview.
Universal, Sony, and Warner just proposed strict new rules to ban AI-generated tracks from official music charts unless they are 'substantially human made.' Under the proposal, songs must also use properly licensed AI models with verified training data rights. Why it matters: The industry is moving to keep AI slop off the hit lists, though defining what counts as human remains tricky.
Google just released Lyria 3.5, its upgraded music generation model rolling out inside Flow Music. The update packs richer melodies, more expressive vocals with cleaner pronunciation, and sharper lyric formatting that actually follows prompts. It also adds direct control over track tempo and song length, giving creators significantly more say over the final track.
A new open-source system called VocalRender generates 48 kHz expressive vocals directly from lyrics and MIDI notes without manual timing alignment. The team released full training code, preprocessing scripts, and two 9.5 GB checkpoints under Apache-2.0. Why it matters: it bypasses tedious phonetic-duration tuning, letting composers render vocal tracks straight from symbolic scores.
Independent developers just launched SoullessMusic and SlopTracker to flag synthetic music on Spotify as the streaming giant lags behind on labeling. The tracked AI accounts rake in an estimated $5.7 million annually, with one virtual singer pulling in $1.5 million alone. Why it matters: everyday listeners and artists are being forced to police streaming slop themselves.
ElevenLabs just rolled out References, a new control layer for Music v2 that lets users upload 10-second to 5-minute audio tracks to steer the style, instrumentation, and vibe of generated songs. You can pair audio with text prompts or let it direct generation alone, with automated copyright checks blocking third-party recordings. It is live now on ElevenMusic, ElevenCreative, and via API.
Sony Music just filed a new copyright lawsuit against AI music startup Udio, alleging the platform illegally scraped more than 30,000 recordings from YouTube to train its models. The label is seeking statutory damages of up to $150,000 per infringed work, arguing that machine-generated tracks threaten to cheapen and drown out original recordings. The move follows a judge's refusal to let Sony append the tracks to its 2024 case.
Suno just had its internal source code leaked by a hacker, exposing that the AI music generator scraped millions of songs and hundreds of thousands of hours of audio from YouTube Music, Deezer, and Genius. The leak directly backs claims made in major record label lawsuits, offering rare, hard evidence of how music AI models get trained.
Mirelo and Kyutai just open-sourced MuScriptor, a new audio-to-MIDI model that handles full, complex music mixes. Instead of tripping over overlapping instruments like older single-track tools, it transcribes entire songs into separate MIDI tracks for vocals, drums, bass, and keys all at once. It's basically a dream shortcut for producers and learners looking to rip clean tracks straight into their DAW.
Researchers just introduced MuSViT, the first foundation vision model trained specifically on sheet music. Pre-trained on 9.7 million pages from IMSLP by reconstructing masked scores, it grasps musical syntax, pitch, and staff layouts, outperforming general vision encoders across score transcription and symbol detection.
