cpd73 wrote: > You can't 'Enable' Essentia from the plugin page, only select if you > want things to be used if enabled. I want to have the plugin talk to the > server to detect if Essentia is enabled, and if high-level analysis > results are there, etc. Then the plugin can control what is shown. Just > not sure how to do that at the mo... Let me try a couple of things on my end (restarting things, various config settings, etc) to eliminate any 'PEBKAC' (https://en.wikipedia.org/wiki/User_error#PEBKAC/PEBCAK) issues.
cpd73 wrote: > Never used SmartMix, before my time :) However, this mixer is meant to > be driven by DSTM - therefore mixes are based upon current queue items, > not a set of attributes. > > And this mixer does not support that, at the moment. Understood, completely. It's your tool for your use case that you've chosen to freely share. I was commenting on what I saw after I took a peek behind the curtains and started to understand how things were connected. I don't currently have the skills to fork what you've done thus far to extend Similarity into SmartMix 2.0. But maybe someone else does or maybe you do, if the idea of creating playlists or influencing playlists based off of specific Musly/Essentia metadata is interesting or exciting. And if not, no complaints here. This is freeware/donation-ware, and I accept everything that comes with such a plugin. cpd73 wrote: > The *only* thing this mixer does with the Essentia high-level attributes > is to find the 4 highest (above 0.8) or lowest (below 0.2) and then > filter tracks on that. i.e. Seed happy=0.95, candidate happy needs to be > 0.7 or higher. Likewise if seed happy=0.19 (so, not happy) candidate > needs to be 0.3 or lower. > > But I find that filters out a -lot- of tracks - so I'm not 100% sure > about this. Please feel free to look at 'check_attribs' in > 'lib/filters.py' and suggest improvements. Thanks for this information. It will help me understand the code as I look at it. I'll now preface any comments I might make in the future with 1) I'm not a math(s) major and 2) I've only taken a basic Python course and I'm not a programmer by training. My comments may be entirely worthless. But like this tool, they're free and you're welcome to disregard them. :) I've also only had an intro to statistics course at university. If 1000 tracks are selected to build the model music style database and then the 4 highest or lowest are used to filter based on a comparison to the seed track, I wonder if we're running into a sampling bias (or some other statistical bias). The 1000 tracks get further split into genres/genre groups based on relative percentage to the existing catalog and that affects the model. I do understand Similarity is pulling from the entire music collection when selecting a song to play and not just the 1000 model tracks. Like you, I only use local files; I do not use any streaming services. Right now Similarity is a bit of a black box to me and it's not clear how much of an effect any one lever/setting has on a mix, so I've chosen to disable most settings and start fairly wide open. Next, I may go to the opposite extreme and use a very narrow range of setting values. As I get a sense for what's happening, I'll know where to go to make adjustments. For instance, I right-clicked on 'Lucretia My Reflection' from Sisters of Mercy and queried for a list of similar tracks. I was -shocked- by some of the results, including a solo track by Fred Schneider (male singer in B-52s) whose voice and music is distinctive and very un-Sisters of Mercy-like, in my mind. So I went into the database and pulled out the metadata values for the two tracks, and, yep, the values for most categories were close-ish that I could understand why both were in the list. cpd73 wrote: > You do know that these Essentia attributes are the probability that a > song matches a model? i.e. happy=0.9 means that there is a 90% -chance- > that it is happy, not that the song is 90% happy. I do fully understand that and appreciate the distinction. The closer to 0 a value, the less likely it is that type; the closer to 1 the more likely it is. The intro to statistics course taught me that much. I do admit that I don't recall everything from the class though. :) ------------------------------------------------------------------------ prabbit's Profile: http://forums.slimdevices.com/member.php?userid=11142 View this thread: http://forums.slimdevices.com/showthread.php?t=115609 _______________________________________________ plugins mailing list plugins@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/plugins
