Vienna Symphonic Library Forum
Forum Statistics

182,891 users have contributed to 42,261 threads and 254,944 posts.

In the past 24 hours, we have 0 new thread(s), 10 new post(s) and 44 new user(s).

  1. Home
  2. Notation Programs & Vienna
  3. Dorico
  4. [FR]Bringing Synchron and VI Pro Players to the Next Level

  • [FR]Bringing Synchron and VI Pro Players to the Next Level

    ThurisazT
    ThurisazT

    Hello VSL Team,

    I have a suggestions about both the Synchron & the VI Pro Players, which will make them more Dorico and, in the future, MuseScore friendly.

    1. As we know the Dimesion Strings is the only library on the market that supports real Divisi (in general I would like to see this library extended with more articulations and effects... I find it the best String library every created, so I would invest in future expansions)

    - I have more experience with VI Pro when it comes to Dimension Strings. The VI Pro is able to handle Divisi, but unfortunately when they contain a single articulation. A single instance of VI Pro player isn't able to handle multiple simultaneous articulations. Most probably the situation with the Synchron Player is the same. Would be really nice if both of them are able to perform multiple articulations, in Divisi, within a single player instance. Possibly this will lead to reducing the RAM and CPU resources needed for multiple instances per articulation.

    2. Some degree of controllable AI, that can be turned On (for the needs of notation) / Off (for DAW).

    - NotePerformer + NPPE is a pretty amazing AI Player, but it's autonomous and there is no way the performance and the humanization to be influenced by the composer in order to let him / her to leave a personal "fingerprint". The only available corrections that can be made are related to EQ, Reverb and Mics (inside NPPE).

    - Orchestral Tools' - Sine Player offers some small of an AI when it comes to the performance of the different samples, which is nice:




    - The AI provides and ability for articulation length control, as well as ability for addition of new articulations that were not recorded, but created out of the pre-existing ones. This will seriously influence the amount of the available articulations and effects compared to the overall library size.

    - The VI Pro offers Trills up to perfect 4th, while Synchron offers it just to major 2nd. This limitation in the Synchron will disappear when there is an AI.

    - The idea I'm having in mind about the AI as part of Synchron and VI Pro players is give us the possibility to tweak some playback aspects and when we are fine with the result to export it as an audio for further post production for final product. In case someone would like to dive deeper and touch every parameter, then he / she should be able to disable the AI.

    Both players currently benefit the best the DAW world, but not enough the notation and software like Dorico.

    Please, consider about improving both players and make them more Dorico, MuseScore, eventually the upcoming Overture 6 and Encore 6.
    Most probably in the future more notation apps like Dorico will appear and surely by the time Cubase will come closer to Dorico in the conception single player - single staff / track. 😊


    Best regards,
    Thurisaz
  • ThurisazT
    ThurisazT

    P.S: Most probably VE Pro will need some improvement in order to handle AI properly when the libraries are hosted by this app?!
  • andiA
    VSL Team
    andiA

    Hello Thurisaz!

    Thanks for your feedback and your suggestions.
    If you want to have different players of one of our (VI or SYNCHRON-ized) Dimension libraries play different articulations at the same time, I recommend having each player on a different channel and triggering a different VI Pro or Synchron Player. The additional player takes very little additional CPU or RAM.

    Best regards,
    Andi

    Vienna Symphonic Library
  • ThurisazT
    ThurisazT

    Hi Andi,

    @andi said:

    Hello Thurisaz!


    Thanks for your feedback and your suggestions.
    If you want to have different players of one of our (VI or SYNCHRON-ized) Dimension libraries play different articulations at the same time, I recommend having each player on a different channel and triggering a different VI Pro or Synchron Player. The additional player takes very little additional CPU or RAM.


    Best regards,
    Andi


    I'm aware that I can use a separate instance of VI Player per articulation on separate channels in Dorico and VE Pro. And this is how I'm using it currently. But there are problems related to the playbak like notes drop-out. Some notes just don't sound during the playback.

    Best regards,
    Thurisaz
  • mducharmeM
    mducharmeM

    Dorico itself is beginning to add NotePerformer-like functionality to add automatic shaping, I think it makes more sense for Dorico to do it rather than have the player do it.

    Dorico currently has no way of triggering sampled trills/tremolos larger than a major second as a limitation of its expression maps. I am hoping they will add this functionality to Dorico in the future.
  • ThurisazT
    ThurisazT

    Hello @mducharme,

    @mducharme said:

    Dorico itself is beginning to add NotePerformer-like functionality to add automatic shaping, I think it makes more sense for Dorico to do it rather than have the player do it.


    Dorico currently has no way of triggering sampled trills/tremolos larger than a major second as a limitation of its expression maps. I am hoping they will add this functionality to Dorico in the future.


    Well, Dorico isn't able to trigger trills larger than a major 2nd, but the VI Pro offers something very smart, which isn't available in the Synchron Player, and I'm wondering why?! The Trills in the VI Pro aren't sampled, but it's an integrated function that uses the long note articulation and turns it into trilled, that sounds pretty naturally because of the round robins. Dorico is able to use this VI Pro function perfectly, so one could create nice sounding trills by using the VI Player and the Articulate Maps, by Symphonic Riot.

    The AI inside the Synchron and VI Pro players is an important improvement for many reasons...

    1. Check te video that I've shared here about the Sine Player, by Orchestral Tools.
    - It automatically controls the sample volume in order to create smoother and more natural legato transition, or switching between articulations.
    2. Dorico would never be able to extend a library articulations and effects list, but the AI inside the player could do it, even without the need of additional samples to be recorded.
    3. Dorico cannot offer volume and timber control for divided strings, but the Player can do it. 😊
    4. Dorico, and any DAW out there offer only controls like CC, KS, PC, ModWheel, Velocity, Tempo... that take effect on the controls that are available in the Player, so the player has the main responsibility for the performance, the DAW, or Dorico are secondary in the scheme.
    - Have you been able to test Iconica Sketch hosted directly by Dorico, and then hosted by NotePerformer + NPPE? Hosted directly this small library sounds totally crappy, but when it's hosted by NPPE... well completely different story. It sounds amazingly well. 😊

    Best wishes,
    Thurisaz

  • mducharmeM
    mducharmeM

    @Thurisaz said:

    Hello @mducharme,

    The AI inside the Synchron and VI Pro players is an important improvement for many reasons...

    1. Check te video that I've shared here about the Sine Player, by Orchestral Tools.
    - It automatically controls the sample volume in order to create smoother and more natural legato transition, or switching between articulations.
    2. Dorico would never be able to extend a library articulations and effects list, but the AI inside the player could do it, even without the need of additional samples to be recorded.
    3. Dorico cannot offer volume and timber control for divided strings, but the Player can do it. 😊
    4. Dorico, and any DAW out there offer only controls like CC, KS, PC, ModWheel, Velocity, Tempo... that take effect on the controls that are available in the Player, so the player has the main responsibility for the performance, the DAW, or Dorico are secondary in the scheme.
    - Have you been able to test Iconica Sketch hosted directly by Dorico, and then hosted by NotePerformer + NPPE? Hosted directly this small library sounds totally crappy, but when it's hosted by NPPE... well completely different story. It sounds amazingly well. 😊

    Best wishes,
    Thurisaz

    Your point #1 - that isn't an AI in SINE Player. It is just a feature. And how do you know that Synchron Player doesn't already have this feature? It does have the capability to crossfade between articulations when you set them up in a stack, similarly to SINE.

    Your point #2 - What does this have to do with AI? Having more articulations and effects is not AI.

    Your point #3 - correct, the player can do it, but how is this AI?

    Your point #4 - also not AI.

    So, I'm not understanding. You are asking for AI, justified by features that aren't AI, and using an example of OT's SINE player that doesn't have AI. Since nothing you are asking for requires AI, what do you want the AI for?
  • ThurisazT
    ThurisazT

    Hi @mducharme ,

    @mducharme said:


    @Thurisaz said:

    Hello @mducharme,


    The AI inside the Synchron and VI Pro players is an important improvement for many reasons...

    1. Check te video that I've shared here about the Sine Player, by Orchestral Tools.
    - It automatically controls the sample volume in order to create smoother and more natural legato transition, or switching between articulations.
    2. Dorico would never be able to extend a library articulations and effects list, but the AI inside the player could do it, even without the need of additional samples to be recorded.
    3. Dorico cannot offer volume and timber control for divided strings, but the Player can do it. 😊
    4. Dorico, and any DAW out there offer only controls like CC, KS, PC, ModWheel, Velocity, Tempo... that take effect on the controls that are available in the Player, so the player has the main responsibility for the performance, the DAW, or Dorico are secondary in the scheme.
    - Have you been able to test Iconica Sketch hosted directly by Dorico, and then hosted by NotePerformer + NPPE? Hosted directly this small library sounds totally crappy, but when it's hosted by NPPE... well completely different story. It sounds amazingly well. 😊

    Best wishes,
    Thurisaz

    Your point #1 - that isn't an AI in SINE Player. It is just a feature. And how do you know that Synchron Player doesn't already have this feature? It does have the capability to crossfade between articulations when you set them up in a stack, similarly to SINE.


    Your point #2 - What does this have to do with AI? Having more articulations and effects is not AI.


    Your point #3 - correct, the player can do it, but how is this AI?


    Your point #4 - also not AI.


    So, I'm not understanding. You are asking for AI, justified by features that aren't AI, and using an example of OT's SINE player that doesn't have AI. Since nothing you are asking for requires AI, what do you want the AI for?

    Well, you are probably not familiar enough what AI is in the area of music and the virtual instruments....

    1. I'm not sure if Synchron Player has an automatic sample volume control, which is available in OT's - Sine. I'm using the Synchron only for percussion, piano and organ. VI Pro most probably doesn't have it, to my ear... Yes, this isn't a complete AI, but could be considered a partial AI integration. And it's something that NotePerformer does.

    2. What do you think NotePerformer and MuseSounds (both AI) do? NotePerformer combines small recorded samples (low quality) library and sample modeling technology where the AI creates the articulations, controls the articulations length, controls the timbral change according to the dynamics, humanizes the playback... The situation with the MuseSounds is similar, just it has much higher quality sounds and probably more recorded articulation that serve as a starting point for the AI to do it's job.

    3. In case of a library that doesn't offer real Divisi, the AI could emulate the whole Divisi behavior, but this can be only done by the Player wich hosts the library.

    4. This point wasn't strictly AI related at all, but I decided to give you an example how the AI improves the performance of a library that sounds terrible hosted directly in Dorico. I woldn't use Iconica Sketch without NP + NPPE. Here NP has limited influence, only on the performance. It can't create new articulations because the library is hosted by external player - Halion Sonic. Still the performance is amazing.

    I hope now everything is clear to you! 😊


    Best wishes,
    Thurisaz
  • mducharmeM
    mducharmeM

    @Thurisaz said:

    Hi @mducharme ,

    Well, you are probably not familiar enough what AI is in the area of music and the virtual instruments....

    1. I'm not sure if Synchron Player has an automatic sample volume control, which is available in OT's - Sine. I'm using the Synchron only for percussion, piano and organ. VI Pro most probably doesn't have it, to my ear... Yes, this isn't a complete AI, but could be considered a partial AI integration. And it's something that NotePerformer does.

    2. What do you think NotePerformer and MuseSounds (both AI) do? NotePerformer combines small recorded samples (low quality) library and sample modeling technology where the AI creates the articulations, controls the articulations length, controls the timbral change according to the dynamics, humanizes the playback... The situation with the MuseSounds is similar, just it has much higher quality sounds and probably more recorded articulation that serve as a starting point for the AI to do it's job.

    3. In case of a library that doesn't offer real Divisi, the AI could emulate the whole Divisi behavior, but this can be only done by the Player wich hosts the library.

    4. This point wasn't strictly AI related at all, but I decided to give you an example how the AI improves the performance of a library that sounds terrible hosted directly in Dorico. I woldn't use Iconica Sketch without NP + NPPE. Here NP has limited influence, only on the performance. It can't create new articulations because the library is hosted by external player - Halion Sonic. Still the performance is amazing.

    I hope now everything is clear to you! 😊


    Best wishes,
    Thurisaz

    Re #1 - no, automatic gain adjustment is not any form of AI, even a light form. If this is AI, then the cruise control in an old car is also AI, and an old fashioned pocket calculator from the 1980s is AI, etc. You can call absolutely everything AI if you wish, it gets rid of the meaning of the word. Automated gain staging doesn't require any intelligence as it can be done with simple calculations.

    Re #2 - NotePerformer is actually mostly synthesized instruments - there are a few samples in there, but most of it is actually synthesis. The woodwinds and brass for instance are entirely synthesized. I think it is also wrong to call all synthesis AI - would you call the synthesizer that Wendy Carlos used to write Switched On Bach in the 1970's "AI"?

    The part of NotePerformer that would be considered AI to some extent is the shaping of musical phrases as that tends to require more intelligence and making educated guesses about the context, and similarly with MuseSounds, but it doesn't sound like you are asking VSL to start adding extra shaping that isn't already there.

    Re #3 - How would you emulate divisi? If the library does not actually have smaller sections sampled, all it could do is reduce the volume to try to simulate divisi. Divisi in sample libraries is generally pretty overrated anyway - the only thing you have to watch out for is heavy divisi with sections that are too large can sound strange.

    Re #4 = The reason Iconica Sketch doesn't sound very good is because of consistency and agility issues that NotePerformer helps to address. This isn't the result of AI but is the direct result of Arne's experiments to determine how to create more agile patches by layering pre-existing articulations (like layering staccato with legato to get a legato with a heavier attack to get runs). This also isn't AI.

    The only thing that you've mentioned that is actually AI is the shaping that is looking ahead at the upcoming music, and both NotePerformer and MuseSounds do this, yes, but I feel this is something that VSL doesn't need to address for the most part as Dorico itself is making improvements in this regard, and VSL building more automatic shaping into the player would result in it sometimes doing things that you don't want as a "black box", since it would be doing extra CC curves that Dorico isn't giving it. It is the problem with using NotePerformer to do a finished product, it is a black box, and if it does something you don't want, you lose the control.
  • R
    R

    In my opinion, the part of AI can only lie on the DAW side, both in terms of "guessing" the length of the samples based on the musical context, and in terms of final mixing, where AI can do a lot.

    In a player like Synchron, frankly, I don't see where AI can be exploited...