I brought it up in the description, but I’ll say it here too. Synthesizer V Studio doesn’t actually officially support Spanish as a language. The reason it’s here is because SOLARIA actually comes with a custom dictionary that allows her to speak Spanish with the ENG->JPN Cross-Lingual Synthesis option. In case you were wondering. Also, unlike the other languages, some phoneme editing was done for Spanish, since it requires some adjustment to work properly. ( Sorry if the trill on “roja” doesn’t sound right. Solaria often struggles when forced to trill her “r”’s while in Japanese XLS )
@@mr.fiffles Among the other extras SOLARIA comes with, it can be found in the Google Drive folder where her installer is hosted. IDK if you still get sent to that Google Drive if you bought her from Dreamtonics, but the link to it should also be in her setup manual.
As a SOLARIA owner, I can say that her Japanese is great. Sure, she has that American accent but that’s fine. Many Japanese singers sing intentionally with very sharp aspirated consonants and very relaxed vowels, look up V-Kei as example. So, her accent can be even favourable for Japanese producers! If her “too aspirated” consonants really bothers, there is a workaround: you can edit her k/t/p sound much shorter and non-aspirated. Change her consonant length from 100% to 20%, you’re welcome.
I agree, I really love her Japanese! When using it, whether intentionally or not, it feels less like a symptom of SOLARIA being an English vocal, and more like a deliberate stylistic choice, and it just makes me love her all that much more
@@YukitoYuki i do not blame these japanese fans her japanese actually feels real since in japanese, consonants are actually slightly aspirated, and most vowels are usually closer(except /o/ and /e/ who are mid instead of close-mid)
Oh yes! She’s my favorite library currently, and I cannot recommend her enough. If you do purchase her though, I recommend downloading her 1.04 version in addition to her most recent update. Reason being that her HDVM-compatible version is a little weird compared to her previous versions. Still good! But it’s best to also have her previous versions too. You can download the previous versions of SynthV’s with a valid serial code here: auth.dreamtonics.com/store/download
@@theamazingkaleb3077 Yes actually! She was made available on the Dreamtonics store a couple(?) of months ago I think, and is $89.99 on there. That said, if you’re asking because of the link I posted, while Dreamtonics has stated that it’s only meant for downloading previous versions of vocals sold on their store, it actually works for virtually any Synthesizer V vocal with a valid serial code.
It’s the default mode, as in, none of the other Vocal Modes are active. Parameters such as Tension, Breathiness, and Pitch Shift, are all also at their default values. Basically this is SOLARIA at her most raw
Perhaps I should further clarify - the vocal modes aren’t “presets” or use in-engine parameters or anything if that’s what you’re thinking. Rather, they are parameters all onto themselves. They’re more like different facets of the sampled singer’s singing. So, for instance, SOLARIA’s “Power” vocal mode is based on her voice provider’s singing when she sings powerfully, if that makes sense.
@@devarni I think that’s more of a compression issue in the samples. The devs probably lowered or normalized the peaks of the vocal modes. It was a particularly known issue with her “Power” Vocal Mode, which has since been… not exactly “fixed”, but it’s not nearly as much of an issue in later versions of her voicebank. This video was made shortly after SOLARIA received vocal modes, so, that’s why you still hear that here.
I am a little offput by her Japanese phonemes, they have a bit too much of the American accent, though I understand that English is her default voicebank and they probably wanted the tone of her voice to stay consistent through languages (or they wanted to save space). Other than that, Solaria is a wonderful voicebank.
Well, I don’t think Eclipsed Sounds really made any adjustments to her sounds for Cross-Lingual Synthesis. Her development was mostly finished by the time Cross-Lingual Synthesis came out, so making new recordings for it probably wasn’t even in consideration. Also considering that the voicer is an American Broadway actress who doesn’t have any kind of known experience with Japanese or Chinese. From what I understand, Cross-Lingual Synthesis is, at least by default when there are no auxiliary recordings, achieved by modifying and re-delegating sounds from the original voicebank phonemes to new phonemes for the intended language. So for example, in Japanese, SOLARIA’s “k a” is a slightly modified version of her “k aa” from her English data ( I don’t necessarily know that it’s exactly that sound, it could be “k ah” or something else ). Some of these are more complex than others, such as achieving an English/Chinese “r” with a Japanese voicebank, which have no such sounds, but that’s the general jist. So it’s partially for this reason that English AI banks tend to have a particularly thick accent when using Cross-Lingual Synthesis to speak Japanese ( Maki English being the only real exception, for potentially obvious reasons ).
her pronunciation is completely understandable so i don't think it's a problem. and considering how other voicebanks have slight accents when not using their original language, it might be on purpose? personally i find it charming
I don’t mind commenters asking questions, but please do not spam the same question verbatim on multiple videos. In any case, to answer your question, it’s kind of complicated. Up until the most recent updates, Vocal Modes have been primarily set per track, though you _could_ also change vocal modes mid-track by joining the notes you wanted changed into a “Group” and changing the Vocal Mode of that specific group. That said, in recent updates, Vocal Modes can now also be used as parameters, so you can draw how prominent they are in the “parameter” panel, not unlike how Cross-Synthesis worked in VOCALOID4.
I do not currently own Rikka unfortunately. Unfortunately, she does not have Vocal Modes yet either. While her companion vocals Natsuki Karin recently received Vocal Modes, and Hanakuma Chifuyu is set to receive them as well, it seems that Rikka requires additional recording for Vocal Modes, as her current training data is not varied enough to implement vocal modes. It is for that reason she will receive a Vocal Mode update at a later date.
@@Ecliptic-P I don’t normally distribute my ust’s/svp’s/vsqx’s/etc. And if I’m being honest, Akatonbo’s a pretty easy song to sequence; it’s basically on the level of a nursery rhyme. I don’t really see a point in distributing it tbh
I brought it up in the description, but I’ll say it here too. Synthesizer V Studio doesn’t actually officially support Spanish as a language. The reason it’s here is because SOLARIA actually comes with a custom dictionary that allows her to speak Spanish with the ENG->JPN Cross-Lingual Synthesis option. In case you were wondering. Also, unlike the other languages, some phoneme editing was done for Spanish, since it requires some adjustment to work properly. ( Sorry if the trill on “roja” doesn’t sound right. Solaria often struggles when forced to trill her “r”’s while in Japanese XLS )
Where am I able to find the Spanish dictionary? is it automatically applied to her voice when it's installed?
@@mr.fiffles Among the other extras SOLARIA comes with, it can be found in the Google Drive folder where her installer is hosted.
IDK if you still get sent to that Google Drive if you bought her from Dreamtonics, but the link to it should also be in her setup manual.
@@dead_byte I did not get that file at all? Can someone give me the actual file?
As a SOLARIA owner, I can say that her Japanese is great.
Sure, she has that American accent but that’s fine. Many Japanese singers sing intentionally with very sharp aspirated consonants and very relaxed vowels, look up V-Kei as example. So, her accent can be even favourable for Japanese producers!
If her “too aspirated” consonants really bothers, there is a workaround: you can edit her k/t/p sound much shorter and non-aspirated. Change her consonant length from 100% to 20%, you’re welcome.
I agree, I really love her Japanese! When using it, whether intentionally or not, it feels less like a symptom of SOLARIA being an English vocal, and more like a deliberate stylistic choice, and it just makes me love her all that much more
@@dead_byte exactly, you worded it perfectly. What I have spoke to many Japanese people and fans, they do adore her Japanese
@@YukitoYuki i do not blame these japanese fans
her japanese actually feels real since in japanese, consonants are actually slightly aspirated, and most vowels are usually closer(except /o/ and /e/ who are mid instead of close-mid)
Muchas gracias por colocar el bonus del español
im getting goosebumps she sounds so reeeAAAALL-
Solaria spanish!!! Yay!!!
Solaria is one of the better-sounding vocal libraries to me personally, I'm messing with her Lite version to learn how to use the software
Oh yes! She’s my favorite library currently, and I cannot recommend her enough.
If you do purchase her though, I recommend downloading her 1.04 version in addition to her most recent update. Reason being that her HDVM-compatible version is a little weird compared to her previous versions. Still good! But it’s best to also have her previous versions too. You can download the previous versions of SynthV’s with a valid serial code here: auth.dreamtonics.com/store/download
@@dead_byte she's also available at dreamthonics store?
@@theamazingkaleb3077 Yes actually! She was made available on the Dreamtonics store a couple(?) of months ago I think, and is $89.99 on there.
That said, if you’re asking because of the link I posted, while Dreamtonics has stated that it’s only meant for downloading previous versions of vocals sold on their store, it actually works for virtually any Synthesizer V vocal with a valid serial code.
@@dead_byte good thing i add all of synth v voice databases and synthesizer v pro studios on my birthday list
This sounds amazing. Which preset does the default mode use or is it a custom mode?
It’s the default mode, as in, none of the other Vocal Modes are active. Parameters such as Tension, Breathiness, and Pitch Shift, are all also at their default values. Basically this is SOLARIA at her most raw
Perhaps I should further clarify - the vocal modes aren’t “presets” or use in-engine parameters or anything if that’s what you’re thinking. Rather, they are parameters all onto themselves. They’re more like different facets of the sampled singer’s singing. So, for instance, SOLARIA’s “Power” vocal mode is based on her voice provider’s singing when she sings powerfully, if that makes sense.
That makes sense. Hence the differences in level? "Soft", for example, is much louder than other modes
@@devarni I think that’s more of a compression issue in the samples. The devs probably lowered or normalized the peaks of the vocal modes. It was a particularly known issue with her “Power” Vocal Mode, which has since been… not exactly “fixed”, but it’s not nearly as much of an issue in later versions of her voicebank. This video was made shortly after SOLARIA received vocal modes, so, that’s why you still hear that here.
I am a little offput by her Japanese phonemes, they have a bit too much of the American accent, though I understand that English is her default voicebank and they probably wanted the tone of her voice to stay consistent through languages (or they wanted to save space). Other than that, Solaria is a wonderful voicebank.
Well, I don’t think Eclipsed Sounds really made any adjustments to her sounds for Cross-Lingual Synthesis. Her development was mostly finished by the time Cross-Lingual Synthesis came out, so making new recordings for it probably wasn’t even in consideration. Also considering that the voicer is an American Broadway actress who doesn’t have any kind of known experience with Japanese or Chinese.
From what I understand, Cross-Lingual Synthesis is, at least by default when there are no auxiliary recordings, achieved by modifying and re-delegating sounds from the original voicebank phonemes to new phonemes for the intended language. So for example, in Japanese, SOLARIA’s “k a” is a slightly modified version of her “k aa” from her English data ( I don’t necessarily know that it’s exactly that sound, it could be “k ah” or something else ). Some of these are more complex than others, such as achieving an English/Chinese “r” with a Japanese voicebank, which have no such sounds, but that’s the general jist. So it’s partially for this reason that English AI banks tend to have a particularly thick accent when using Cross-Lingual Synthesis to speak Japanese ( Maki English being the only real exception, for potentially obvious reasons ).
私は日本人ですが、
J-POPアーティストの中にはこのような発音で歌う女性が数多くいるので、
あまり違和感を感じませんでした。
her pronunciation is completely understandable so i don't think it's a problem. and considering how other voicebanks have slight accents when not using their original language, it might be on purpose? personally i find it charming
all other modes: normal
powerful: [distant yelling]
i only recently got synth v pro (and solaria) but she doesn't have the vocal modes! how do i update to get them? :0
Does it say version 103?
@@corasundae yeah turns out i just had to click update in the licensing tab!!
@@miksnails Glad you got it working!
Can't wait to have all synth v voice databases and synth v pro for my birthday
I really don't like when it comes to the power vocal mode and the voice sudden become distanced
yeah like too much reverb
can we select different vocal modes for different notes & words? or is it applied universally to the entire track?
I don’t mind commenters asking questions, but please do not spam the same question verbatim on multiple videos.
In any case, to answer your question, it’s kind of complicated. Up until the most recent updates, Vocal Modes have been primarily set per track, though you _could_ also change vocal modes mid-track by joining the notes you wanted changed into a “Group” and changing the Vocal Mode of that specific group.
That said, in recent updates, Vocal Modes can now also be used as parameters, so you can draw how prominent they are in the “parameter” panel, not unlike how Cross-Synthesis worked in VOCALOID4.
Do you have Rikka? and if so, do you plan to do a video with her vocal modes soon?
I do not currently own Rikka unfortunately. Unfortunately, she does not have Vocal Modes yet either. While her companion vocals Natsuki Karin recently received Vocal Modes, and Hanakuma Chifuyu is set to receive them as well, it seems that Rikka requires additional recording for Vocal Modes, as her current training data is not varied enough to implement vocal modes. It is for that reason she will receive a Vocal Mode update at a later date.
where did you get the svp for the test song?
I made it myself
@@dead_byte can you release it?
@@Ecliptic-P I don’t normally distribute my ust’s/svp’s/vsqx’s/etc.
And if I’m being honest, Akatonbo’s a pretty easy song to sequence; it’s basically on the level of a nursery rhyme. I don’t really see a point in distributing it tbh
@@dead_byte I’m lazy XD