locrian Posted December 12, 2023 Share Posted December 12, 2023 (edited) Review: https://bedroomproducersblog.com/2023/12/12/output-pack-generator/ D/L: https://coproducer.output.com/ Output released Pack Generator, a free AI-based sound generator that lets users create royalty-free sample packs using text prompts. Edited December 12, 2023 by locrian Link to comment Share on other sites More sharing options...
locrian Posted December 12, 2023 Author Share Posted December 12, 2023 (edited) nm Edited December 12, 2023 by locrian Link to comment Share on other sites More sharing options...
user 905133 Posted December 12, 2023 Share Posted December 12, 2023 Hmmmm. I tried "one to two minute slowly evolving, pleasantly mellow, ambient textures." I am not trying to figure out what words I can use to get what I was looking for. I had in mind sounds that were from one to two minutes in length each where a single sound slowly evolved in harmonic content (i.e., texturally) from the start of the sound to the end. I think I understand how results I got are related to some of the key words I used but are not what I thought I was asking for. Link to comment Share on other sites More sharing options...
Lionel Posted December 12, 2023 Share Posted December 12, 2023 Very interesting. I very much see this as competition to human-created content. Skynet’s coming for us all, hah 2 Link to comment Share on other sites More sharing options...
GoncaloL Posted December 12, 2023 Share Posted December 12, 2023 I did let my app working hard on beats and melodies…and i came out for a drink. Already told it that we need that album ready early next year 1 Link to comment Share on other sites More sharing options...
user 905133 Posted December 12, 2023 Share Posted December 12, 2023 (edited) So, I listened to the demo tunes for first set of 4 "packs" and had "it" create 4 more using the same prompt (unchanged). Then I started listening to the wav files that went into making the demo tunes for some of the packs. I thought of how I could try to reword the prompt to get closer to what I want (such as adding "using high quality non-orchestral sounding audio samples"), but I decided (1) I don't want to continue trying to train the software by using the thumb up/thumb down buttons for each set of 4 sample packs based on the demo tunes, (2) I sincerely believe it would take far less time to create my own layers to combine without sounding like garbage, (3) I am starting to get a nasty headache from listening to the demo tunes and the wav files trying to figure out how to reword my request to try to come closer to what I thought I was describing. 1 hour ago, Lionel said: Very interesting. I very much see this as competition to human-created content. For me it might be more interesting (1) if the "try to read my mind based on the words I used" game came closer to what I had in mind (and have done with my own) and (2) if I had more time and interest to try to train the software. As for being competitive, I suspect this kind of thing can become more "cost effective" that using human to create content. Edited December 12, 2023 by User 905133 deleted stray word (x 2) 2 Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now