I’m with you on that one... Haiyi has a LOT of power in her raw state and can be pushed even further with enough editing. After hearing her soft and sweet demos, I was shocked when I used her for the first time and she just screamed everything.
SynthV Medium⁵ Updates/Discussion Thread (Quadimension voices)
- Thread starter xuu
- Start date
And this isn't a fact but I and my gf are both speculating that at least one pitch from her power bank could have been used in the SynthV bank because she has noticeable switch in tone in her upper range that has a lot of power to it.
Also I just noticed that UTAU Haiyi had vocal fry recordings
ah I wish they kept those
Also I just noticed that UTAU Haiyi had vocal fry recordings
ah I wish they kept those
This thread has been collecting dust wwww
Btw just checked Dreamtonics' Bilibili and in Gen 3 SynthV AI announcement video Ddickky commented this:
Btw just checked Dreamtonics' Bilibili and in Gen 3 SynthV AI announcement video Ddickky commented this:
Anybody understands Chinese here? Could you clarify what he's talking about?
Seems he talks about how he wanted Stardust/Xingchen to be the first AI library (or be among the first ones) among the MEDIUM5 crew (?)... And that because of known reasons, it's a bit of a struggle to do so..... I'd like to wait for somebody who knows it better because it sounds a bit vague in some places
I haven't been in the Chinese loop for a while. Is zero still a thing and is zero English still a thing? What's the difference between minis zero and stardust? Is stardust rights still being hold up by vocaloid?
About banks... I don't know. But on the lore side of things it is known that all the quad girls came from a single being called ZERO, which split into them, and what basically was the leftovers turned into Minus. Considering Stardust is a VOCALOID4, Yamaha probably still has rights to her name.
If memory serves, Yamaha’s got nothing to do with it; Quadimension made Xingchen through another company, I think, and Chalili’s contract still hasn’t expired. They could, in theory, make Xingchen with a new VP (which was sort of the idea behind Minus), but they haven’t for reasons yet to be explained.
ZERO has basically turned back into just being a concept I think. Since AI is now possible there's no reason to mess around with the Xingchen/ZERO thing they would have had to do with a Standard voice being compared to VOCALOID4 Xingchen. Xingchen's issue is that there are unresolved contractual issues with Shanghai HENIAN who were responsible for developing her V4 library, but they seem to be getting close to a resolution.
[Edited out]
EDIT: My friend pointed out to me that Ddickky posted ~3 more posts before this one, which changes things a bit. I'll try to summarise. Please bear in mind I don't speak Chinese very well at all and I'm relying on cross-referencing through multiple softwares and dictionaries. Do note that Ddickky is mostly talking about V4 era here since that was when he was involved with Vocaloid.
Post 1 - Even now he still sees comments surprised by Synthesizer V's performance in the engine market. He mentions being so "conservative" in business that his friends said he would miss big opportunities, but he says he knew exactly what he wanted, saying it's okay to take things slow and steady since he's not in a hurry. He says the perception of companies who've had experience of multiple engines is probably better than that of those who know only one. He says Vocaloid's trajectory was set at the end of 2014 (V4 release), and Crypton beginning to research for NT was like the beginning of the end.
Post 2 - A decade ago, Vocaloid dominated the vocal synth industry through technology alone. Third-parties and new companies, even the likes of Crypton, had no choice but to go with Yamaha. In addition to the considerable cost of producing a sound library, you had to pay Yamaha part of the sales and it was very hard to make any money. You also needed the VOCALOID Editor to use a voice library, so for every new person buying a library there was a sale for the editor, which meant while the company earnt 500 (yuan) Yamaha would earn 600. Yamaha was very satisfied and even though companies wanted to do something about it they couldn't. Crypton eventually created their own editor and it was very successful.
Post 3 - A healthy up-and-down relationship (first and third party) should be one that builds deeper connections, mutual understanding and care, and reduces gaming/messing around to increase efficiency. When Synthesizer V started, Ddickky invested a lot and made four libraries in the first year. Quadimension didn't need to be so urgent, but it was important to help Dreamtonics. Industry recognition of Synthesizer V needed to be established quickly and they contributed a lot to Dreamtonics' revenue that year. Most people are not forward-looking and like to go with the crowd. He says Synthesizer V R1's performance and business relations were already outperforming Vocaloid's. Perhaps Japanese companies were more cautious and wanted to wait and see. Back then Kanru and Ddickky were in a hurry to get it out there, but now that they have AHS to help Ddickky is less anxious. He says Dreamtonics will always have their support and they're happy to back them up whenever needed.
Post 4 - Ddickky starts by noting they haven't promoted SynthV AI yet, and says that a little selfishly he wants to have the first AI library from Quadimension be one called Xingchen Infinity, but because of well-known reasons there's some trouble with this. He's still thinking of a solution. At the end of the month they'll be in contact with Chalili apparently. He says, observe fate, everyone. He believes in fate now, karma, it's too wonderful for words.
The last part makes a lot more sense now. He essentially gave us the whole story of why they stopped doing Vocaloid and moved to SynthV, while also praising Kanru for how he's handled things so far. Looks like we might be getting more voices from Quadimension soon!
[Edited out]
EDIT: My friend pointed out to me that Ddickky posted ~3 more posts before this one, which changes things a bit. I'll try to summarise. Please bear in mind I don't speak Chinese very well at all and I'm relying on cross-referencing through multiple softwares and dictionaries. Do note that Ddickky is mostly talking about V4 era here since that was when he was involved with Vocaloid.
Post 1 - Even now he still sees comments surprised by Synthesizer V's performance in the engine market. He mentions being so "conservative" in business that his friends said he would miss big opportunities, but he says he knew exactly what he wanted, saying it's okay to take things slow and steady since he's not in a hurry. He says the perception of companies who've had experience of multiple engines is probably better than that of those who know only one. He says Vocaloid's trajectory was set at the end of 2014 (V4 release), and Crypton beginning to research for NT was like the beginning of the end.
Post 2 - A decade ago, Vocaloid dominated the vocal synth industry through technology alone. Third-parties and new companies, even the likes of Crypton, had no choice but to go with Yamaha. In addition to the considerable cost of producing a sound library, you had to pay Yamaha part of the sales and it was very hard to make any money. You also needed the VOCALOID Editor to use a voice library, so for every new person buying a library there was a sale for the editor, which meant while the company earnt 500 (yuan) Yamaha would earn 600. Yamaha was very satisfied and even though companies wanted to do something about it they couldn't. Crypton eventually created their own editor and it was very successful.
Post 3 - A healthy up-and-down relationship (first and third party) should be one that builds deeper connections, mutual understanding and care, and reduces gaming/messing around to increase efficiency. When Synthesizer V started, Ddickky invested a lot and made four libraries in the first year. Quadimension didn't need to be so urgent, but it was important to help Dreamtonics. Industry recognition of Synthesizer V needed to be established quickly and they contributed a lot to Dreamtonics' revenue that year. Most people are not forward-looking and like to go with the crowd. He says Synthesizer V R1's performance and business relations were already outperforming Vocaloid's. Perhaps Japanese companies were more cautious and wanted to wait and see. Back then Kanru and Ddickky were in a hurry to get it out there, but now that they have AHS to help Ddickky is less anxious. He says Dreamtonics will always have their support and they're happy to back them up whenever needed.
Post 4 - Ddickky starts by noting they haven't promoted SynthV AI yet, and says that a little selfishly he wants to have the first AI library from Quadimension be one called Xingchen Infinity, but because of well-known reasons there's some trouble with this. He's still thinking of a solution. At the end of the month they'll be in contact with Chalili apparently. He says, observe fate, everyone. He believes in fate now, karma, it's too wonderful for words.
The last part makes a lot more sense now. He essentially gave us the whole story of why they stopped doing Vocaloid and moved to SynthV, while also praising Kanru for how he's handled things so far. Looks like we might be getting more voices from Quadimension soon!
Last edited:
I mean, it's deserved. Synthesizer V has been doing business relations super well, and Kanru really struck gold with the model he made. We can only hope for more Studio vocals, and good wishes for R3!
To me it sounds like the exact same vb just re-tuned, I'd be damned if it's a new one haha I think people keep forgetting just how soft you can actually go with Chiyu and how good she really is
There is a Quadimension thread! I was certain of it but it was hard to look up on mobile orz. Anyway, in addition to all the Synth V news today, there's also some big news related to the Quadimension girls:
I'm excited, but I hope it's not the same as how they did the transition to R2 and are making the Standard voices unusable anymore if you activate the AI banks orz
I'm excited, but I hope it's not the same as how they did the transition to R2 and are making the Standard voices unusable anymore if you activate the AI banks orz
Don't worry, they will NOT deactivate the standard voices. In fact, they're co-existing with the AI voices.
Here's what Cangqiong ai english sounds like
www.bilibili.com
【苍穹AI英语内测版】孤勇者【抢鲜版】_哔哩哔哩_bilibili
只能说SV太强了，赶紧整一段。完整版会配个牧心来对唱~（内侧声库，非最终效果）歌曲：孤勇者原唱：陈奕迅作曲：钱雷编曲：钱雷英文填词：肖恩Shaun-Gibson画师：原子Dan翻唱：苍穹AI 英文内测版调校：瑞安Ryan不跑调, 视频播放量 82191、弹幕量 152、点赞数 7262、投硬币枚数 2064、收藏人数 3410、转发人数 1079, 视频作者 瑞安Ryan不跑调, 作者简介 用爱发电，为爱发癫，相关视频：【苍穹×墨清弦】Magnet【跨引擎CP】，【苍穹、赤羽、星尘】你醒啦？说，我是不是《你姐》？，【牧心/牧馨】日不落【无敌美少女】，“毕业那天一定要杀光你们。”...
Another Cangqiong ai english sample I don't know how to upgrade her bank or if it's even out but still
Sorry for the triple post but more info came out Haiyi's AI is in development and Cangqiong is in beta and kanru is having a well deserve vacation
Last edited: