(Edit. It’s not a good idea because of programming)
Many users have reported that the new engine makes pre-Weina Voicebanks sound less clean and with messy pronunciation/expressions. That’s why I personally consider more appropriate to go back to the previous version while maintaining 1.7.0 just for Weina (Maybe continuing working on the new HDVM engine to fully make it compatible with all voicebanks)
We know that Weina is a voicebank designed to be superior than others, since it has a higher price, being part of the Gold Series and having as a voice provider a famous singer.
So it might be a good idea to give only Weina her own engine, precisely version 1.7.0.
I mean, I think it’s a good idea for Gold Series voicebanks to have their own specific engines for each voice. That would allow to adjust and improve the engine to the specific characteristics of the voice provider instead of always designing the engine for a variety of different voices in a generic and open way.
This would maintain the quality of the other older voicebanks and at the same time enhance the desirability of the Gold Series by giving them more control of use and engine specificity. Therefore more personality and uniqueness.
(I don’t know programming and I don’t fully understand how an AI works either. But I think this would make more users happy and would also give Weina more unique features that could attract more customers)
(Please correct me if any of this information is wrong)
I am having MASSIVE problems. All my songs all of a sudden sound like crap. Is there an installer for 1.6.1 somewhere : I desperately need to go back!!!
I don’t really know. At least I know companies are giving support to install the old versions of the voicebanks. But I’m not sure if Synth V itself has a link to the older version. I tried restoring my PC to a restore point and it worked for the Synth.
Installers for previous versions are downloaded to Documents\Synthesizer V Downloads (on Windows) and are not cleaned up after the update is installed, so check if you still have the 1.6.1 installer there. Keep in mind this only applies to the SynthV Studio installer, voice installers are not kept in the same way.
Unfortuantely maintaining multiple code lines in this way is a software development antipattern. Having a single engine with the ability to toggle on and off certain features would be the ideal situation, in that it allows the user more control over the functionality but also makes future development and maintenance easier for Dreamtonics.
Edit: it’s also worth noting that while some people are frustrated with the inability to completely disable HDVM, it isn’t behaving poorly for all non-Gold voices. Some like Karin and Maki are having significant issues, but I’ve heard Solaria and Eleanor have responded relatively well to the new feature (though this may vary quite a bit depending on the specific SVP and I definitely agree that it’s a big problem if users cannot produce the same audio output as with previous versions due to the new features being forcibly enabled, even if they generally sound “better” than before).
Hi friends, put down your torches, calm down and be patient, Kanru will do whatever it takes, I trust him
I also find there is a problem, I miss my Kevin and my Solaria but I’m patient, I’m sure an update will come soon
After four hours of nerve wrecking searching via TimeMachine ( on Mac) I was able to reconstruct 1.6.1 and the application support folder with it’s databases, Now I have only been able to check this on one song , but that sounded “normal” again. No raspy voice and strange double deep vibrato dips ( which sounded terrible) also no more NOISE! That was the worst bit. It was like switching from Eleanor AI to Eleanor Lite!! Please : whatever you do Dreamtonics incorporate a way to disable that new AI modus created for Weina and the. new HIghDynamics voices. Or make sure the old voices ( I HAD updated all to the latest version for 1.7) I do find SynthV the Master of Vocal Synthesis and LOVE the product, and will invest in new English voices( I have all there are right now but I hope Dreamtonics can help me keep a consistent playback between versions.
Even if making a separate engine per voice would be a bit much, I see where you’re coming from. Making SynthV Gold its own engine would have been a good idea, and just run Synthesizer V Studio and Synthesizer V Gold as sister brands like CeVIO AI and Voisona.
Solaria and Eleanor (the AI versions) may work relatively well, but I still had to fix LOTS of words after I had to re-render the vocals a after a lyrics change.
Words like “drinks” or “searching” are almost impossible to make them sound right after the update.
As a paying customer I’m disappointed now!
but I hope Dreamtonics can help me keep a consistent playback between versions.