Fer Isella

Great news! limbo/ has a direct deal with @qobuz 🔊 The streaming platform built for listeners who truly value music and sound, Qobuz offers lossless high-resolution audio, up to 24-bit quality, and a catalog of more than 100 million tracks, all curated by real music experts.

limbomusic's tweet photo. Great news! limbo/ has a direct deal with @qobuz 🔊

The streaming platform built for listeners who truly value music and sound, Qobuz offers lossless high-resolution audio, up to 24-bit quality, and a catalog of more than 100 million tracks, all curated by real music experts. https://t.co/3XshJlBcc5

Fer Isella

@FerIsella

about 16 hours ago

@UnamunoAgain amo

Fer Isella

@FerIsella

about 16 hours ago

@TrollFootball 🖕

Fer Isella

@FerIsella

about 16 hours ago

@Joi2James agree

Fer Isella

@FerIsella

about 16 hours ago

@harleyf https://t.co/QXVa4vLG52 😍

453

Fer Isella

@FerIsella

about 17 hours ago

@TTrimoreau Audience

Fer Isella

@FerIsella

about 17 hours ago

Suno is "exploring a developer API." The pitch: a curated group of partners building applications "that unlock experiences generative music makes possible for the first time." Let me describe one of those experiences. Streaming royalties are still mostly pro-rata: one pool of money, divided by total streams. Every synthetic track that captures plays takes its fraction from the same pool that pays human artists. People uploading AI tracks one at a time already do this. An API lets you do it at industrial volume, programmatically, from inside somebody else's product. Some context worth keeping in view. Universal and Sony are suing Suno over the recordings used to train these models. In May, after fingerprinting Suno's training data, they asked the court to expand the case to more than 61,000 identified works. A summary judgment hearing on the fair use question is expected this month. Suno's response so far: raise 400 million dollars and start recruiting developers. Build the company on material the courts haven't ruled on. Open the platform before they do. We've spent twenty years building music technology. At @limbomusic/, we work with AI every day to protect catalogs, not to replace the people who made them. So this is not fear of the tools. It's an objection to one specific business model: take value from human work without consent, then scale the extraction while its legality is still an open question in court. Somebody's years are inside every song in that training data. Lessons, rent, failed bands, the whole cost of a life in music. When a fraction of a cent moves from that person to a synthetic track, that movement is not weather. Someone designed it. Most of the industry is applauding. So, honest question: are we not seeing it, or do we see it and think it's fine? I don't. #Independent in a moment of concentration. #Human in a moment when not everything that sounds like music is.

$FerIsella's tweet photo. Suno is "exploring a developer API." The pitch: a curated group of partners building applications "that unlock experiences generative music makes possible for the first time." Let me describe one of those experiences. Streaming royalties are still mostly pro-rata: one pool of money, divided by total streams. Every synthetic track that captures plays takes its fraction from the same pool that pays human artists. People uploading AI tracks one at a time already do this. An API lets you do it at industrial volume, programmatically, from inside somebody else's product. Some context worth keeping in view. Universal and Sony are suing Suno over the recordings used to train these models. In May, after fingerprinting Suno's training data, they asked the court to expand the case to more than 61,000 identified works. A summary judgment hearing on the fair use question is expected this month. Suno's response so far: raise 400 million dollars and start recruiting developers. Build the company on material the courts haven't ruled on. Open the platform before they do. We've spent twenty years building music technology. At @limbomusic/, we work with AI every day to protect catalogs, not to replace the people who made them. So this is not fear of the tools. It's an objection to one specific business model: take value from human work without consent, then scale the extraction while its legality is still an open question in court. Somebody's years are inside every song in that training data. Lessons, rent, failed bands, the whole cost of a life in music. When a fraction of a cent moves from that person to a synthetic track, that movement is not weather. Someone designed it. Most of the industry is applauding. So, honest question: are we not seeing it, or do we see it and think it's fine? I don't. #Independent in a moment of concentration. #Human in a moment when not everything that sounds like music is.$