Does anyone have any experience working with Dreamtonics Synthesizer V vocaloid plugin with Dorico? Its kinda shockingly ok for demos but I wish there was a nicer way to combine with Dorico (or any other DAW for that matter).
Synthesizer V lets you import midi and midi with lyrics, which I can’t get Dorico to export.
It also doesn’t seem to import dynamics or expression, unfortunately.
This would be a pretty incredible combo, if it worked somehow together for vocal demos.
For now you have to export midi from Dorico, import it into Synthesizer V and then manually enter in all dynamics, lyrics and anything else to make it less robotic, but the results are great for demos.
'tis the way it’s always been with Vocaloid. I seem to remember that there was a VST with Vocaloid II, but it didn’t work very well.
To answer the question, no I haven’t; I used to do the export routine with Finale. Finale > Sonar, then Sonar MIDI > Vocaloid, then taking the audio file generated by Vocaloid and importing it to the Sonar file. I think I did the same thing once or twice with Dorico – same process, but with Cubase; I only have version III of Vocaloid.
I’m not sure if the dreamtonic version is the same as the Yamaha.
This one works surprisingly well. The interface is a bit of a mess and documentation pretty weak but its surprisingly not bad as an AI corrected singing voice. There is only one of their ‘voices’ that’s set for English, Kevin, and it does fairly well and is fairly programmable. I think, if I could get past the interface and lack of manual…
If it’s backwards compatible, there are a few other English voices around. In v.3 I’ve got Lily, BigAl, Prima, Bruno and Clara. Prima (originally in v.2) was supposed to be an operatic soprano, which is why I bought it. One of my more disappointing purchases.
I took a quick look at the Yamaha version, it does look different from the Dreamtonics one, including different ‘singer’ profiles.
But similar issues, you have to import midi files. But midi files with lyrics embedded show up with the lyrics, which makes it a bit faster. Too bad Dorico can’t export lyrics in a midi export.
Picking up the trail of an old thread here. I’m writing an LGBTQ theater piece for a theater in Kansas City, MO. And Synthesizer V — for the first time in my experience — is working as a VST plug-in within Dorico to allow me to create a fully-sung demo for the client. And to my ear, it’s pretty damn decent. It’s quite easy to use and to integrate. Dorico, of course, handles it like a champ.
This is a brief snippet:
Voice libraries involved:
Ninezero [Dreamtonics]
Jun [Audiologie]
Asterian [Eclipsed Sounds]
Kevin [Dreamtonics]
These were all available thru the Dreamtonics site, and were very reasonably priced.
Your demo is very interesting. Perhaps mixing the voices helps make the pronunciation more Western than many of the Japanese voices by themselves.
As with VOCALOID and EmVoice, there seems to be a challenge finding voices that represent not just different ranges but also different timbres to differentiate heroic tenor from older baritone.
Synthesizer V, like the other vocal simulation programs, seems to be making great strides, but much will depend on evolving localized vowel pronunciations and more diverse ranges/timbres.
What was interesting to me was the variance in vocal timbres available thru Dreamtonics. “Asterian,” for example, is billed as a bass-baritone, and soloed up, the voice quality actually does bear that out. The other three easily flex between high baritone and high tenor. In mixing them together for a TTBB choral project like this, I’m getting surprising results.
And Asterian is also natively English… And so are Nonzero and Kevin, if I’m not mistaken. So they shouldn’t need much massaging.
Quick question: How did you set this up? 4 tracks in SynthV, then record from Dorico one by one? Or 4 instances of the Plugin?
I still have major usability issues using SynthV, as in I can’t set MIDI channels for example.
Precious little time, Dan. To be honest, that audio snippet was my very first attempt, and took maybe 10-15 minutes to assemble, including the instrumental. I just plowed into the app and figured my way out from there.
I can’t seem to find an actual manual for this plugin, but the Dreamtonics YouTube site has extensive - and very helpful – tutorials. Almost all I know came from there.
I haven’t yet figured out how to get the plugin to read notes on Dorico staves, or if that’s even possible. But I created a single vocal part – shockingly easy – and duplicated it for three other voice-library parts. Using the “AI Retake” function I got the app to slightly alter the three copied takes to make it a little more human sounding.
So: Dorico isn’t actually playing the notes in the plugin. The plugin runs concurrently with Dorico, and the notation staves themselves are playback-suppressed to keep the default synth in the plugin from sounding at the same time.
For this snippet, all the voices are routed to a single channel which comes up in the Dorico mixer. I think for my TTBB piece I’ll create four instances of the Synth V plugin, one for each section. That will bring them up in the mixer as individual channels which I can automate in the Key Editor. Likewise for solos, etc.
Do watch the YouTube videos. I found them truly jaw-dropping.
I’ve been playing around with this for some time now, but the clunkiness of the UI was always a little off-putting…
You can get the Plugin to read the music from Dorico, but you have to record the different singers one at a time, and then you render the voices.
If SynthV was able to record different MIDI channels at the same time, it would be easier and faster and keep all the voices in one instance.
I’ve said it before, it’s so tantalizing to know that MIDI is able to transmit lyrics to plugins, but neither Dorico nor SynthV seem to make use of this facility.
I’m actually willing to invest more in their ecosystem, but there’s more kinks to be ironed out at this time.
Thx guys!
The thing is, you need to actually use the record function of the Plugin, and let the Dorico piece run through a many times as you have singers, muting the tracks you don’t want to record.
The problem is, SynthV doesn’t have an auto-record function, so you have to start Dorico’s playback, frantically activate recording in the Plugin, and hope for the best…
The playback/recording controls disappearing all the time doesn’t help matters either…
With multiple instances, you could likely get rid of the muting business and just do as many recording passes as you have singers/instances.
I just found from another YouTube video that clicking inside the pane where the note-boxes appear makes the record/play panel reappear. Once Dorico is playing you’ll have to click the record button twice to pop it into record mode, but then it’s smooth sailing from there.
Correct. But I could never understand why the Plugin needs 2 control sets in the first place, why you do have to click twice to start recording, and so on.
But I’m willing to give them the benefit of the doubt, since the quality is so awesome.
And I think Kanru hasn’t yet fully grasped how large the market for this running inside a notation package truly is.
