stoken6 Posted June 23, 2023 Share Posted June 23, 2023 The latest 2023.6 update of Sibelius include AI-powered chord symbol detection: https://www.avid.com/resource-center/whats-new-in-sibelius-june-2023?utm_campaign=sibelius_2023_6--renew&utm_source=gainsight&utm_medium=email Personally I'm a little underwhelmed. It seems everything needs to be "now with added AI", and the training set is Bach, Beethoven, Haydn, and other material that doesn't really benefit from chord symbols. And the range of chords identified is restricted (nothing more than a seventh). More interestingly, they've deliberately chosen a model that runs locally (no data harvesting), and their blog post (https://www.avid.com/resource-center/data-driven-workflows-within-sibelius) states: "we are [...] committed to using ethical AI practices to respect creators’ data." I wonder how long that will last... Cheers, Mike. Quote Link to comment Share on other sites More sharing options...
ElmerJFudd Posted June 23, 2023 Share Posted June 23, 2023 The step that will really be a massive change in use of AI in music is when the machine has read and listened to enough scores to accurately transcribe music from a recording (right time sig, tempo, key, enharmonic spellings, chord symbols, and instrumentation, etc.). And conversely play back a score from sheet music as a human would. Improvise an arrangement from a melody as a human would including a solo section. That’s the singularity to me for a music AI. I think we are still some years off from this. 1 Quote Yamaha CP88, Casio PX-560 Link to comment Share on other sites More sharing options...
Josh Paxton Posted June 23, 2023 Share Posted June 23, 2023 "AI-powered" will be to the 2020s what the "e-" prefix was to the 1990s. 2 Quote Link to comment Share on other sites More sharing options...
Docbop Posted June 23, 2023 Share Posted June 23, 2023 Going to be a decade of lawsuits and probably new definition of copyright because most of AI is building from knowledge based on material owned by other sources they scrape off the internet. This is what all the current Reddit stink is all about. Company who originally collected the data want $$$ for use of their data collection. A lot is our personal data collected, but everyone else gets to make money off our personal data but us. 1 Quote Link to comment Share on other sites More sharing options...
ElmerJFudd Posted June 23, 2023 Share Posted June 23, 2023 40 minutes ago, Docbop said: Going to be a decade of lawsuits and probably new definition of copyright because most of AI is building from knowledge based on material owned by other sources they scrape off the internet. This is what all the current Reddit stink is all about. Company who originally collected the data want $$$ for use of their data collection. A lot is our personal data collected, but everyone else gets to make money off our personal data but us. They need to have results spit out citations at the end. Quote Yamaha CP88, Casio PX-560 Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.