![]() ![]() I mean, you DO want to see duplicates when you want to get rid of them. I have a couple of questions regarding this though, because it occurs to me that sometimes you DO want to show duplicates and sometimes not. but of course, Sononym is uniquely suited to detecting duplicates - even across sample formats, bit-rates etc. this is something we didn't even consider. Maybe we can come up with a less hackish, integrated solution.Īs Sononym-dev, I agree with this but also want to clarify a bit: We are a bit wary of recommending this trick (symlinking), but yeah - it works At least, that's how we expect this feedback cycle to work.Īnd yes, that would include the ability to "correct" misclassified sounds, somehow. With similarity search being the core feature, everything surrounding it will (hopefully) reflect the features that people actually need in the upcoming versions. That said, they all LOVED similarity search, since it's totally able to capture the character of such abstract sounds, and enables you to explore how they are related to one another. #Sononym sample browse pro#Furthermore, in professional sound libraries (and I mean the PRO level ones), the metadata is often very detailed and therefore, also useful. In sound design, sounds are often quite abstract and therefore very hard to classify. Having said that none of the three are perfect and its worth checking them all out.As Sononym-dev, I agree with this but also want to clarify a bit:ĭuring development we got input from a couple of heavy-weight game audio professionals, and they politely pointed out that the lack of metadata editing would be a big issue for them. Strangely enough the same guy did a video and at the end compares the two, a fair comparison I think.Īlso not being able to recategorise a mislabeled sample is a big flaw. As of now its not a complete solution, however I think its a great product for what it does. So as clever as Sononym is, and it is pretty smart for getting like sounds, it doesn't do the actual basics as good. Now this is a problem because you can't user re-tag them !! Now another sample management program like Samplism will tag by name, certainly not as techie but in this scenario works way better because the sample creators have called there files sensible names that include the relevant keywords like xxx Kick 021 etc. Sononym categorises/tags by their sonic finger print so unfortunately it misses a few, kicks as toms, plucked synth as hihats etc. Say for example you want to find all the kicks in a Splice folder, there could be thousands. Its touts itself as a Sample management program. The problem I found with Sononym is simply this. I found a really nice example video for this: I think you can't completly compare Audiofinder/Samplism with Sononym!īoth manage samples and tag them, but Sononyms strength is finding similar samples #Sononym sample browse mac#Wish them all the best and if not on a Mac its worth a look IMHO Having said that none of the three are perfect and its worth checking them all out. It also uses a lot of CPU in comparison to others.Īlso not being able to recategorise a mislabeled sample is a big flaw. Of course they can't do the spectral matching. Also most sample libraries do tag their samples by name so in the long run using a name based tagging was actually more accurate than Sononym. ![]() I chose Samplism because its very fast but Mac only. I think Sononym shows promise and the idea is good but it's not quite right, right now.Yes I agree with all your points, I tested out the 3 leading contenders Sononym,Audiofinder and Samplism. The ability to add user tags might help with overall management IMO as, right now, I'd probably have to use it alongside Samplism or Audiofinder. It's in the ballpark - give it a hi-hat and it will find high-pitched noisy things with a short decay but the sounds with high overall similarity are quite different sonically and tend to lose out to a search on a decently tagged library in something more conventional. However, I'm finding the app's view of "similarity" to not be that great. But moving it to a local SSD using a symbolic link is much, much faster - but that's not very manageable for multiple libraries. Similarity searches on the NAS-based database take forever and were practically unusable. One thing the developers should add is the ability to move or shadow the database to local storage such as an SSD. ![]() I've tried this on a library of around 100k percussion samples on a NAS. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |