First of all, welcome; I also conducted some experiments with a horrific result.
lupusniger wrote:
3)does anyone know how to write, using votox, the following phrase:NIHIL ME MULCERE POTEST MEUS EST SINE FINE DOLOR
On this, I can tell you that you must pretend that the reader is an English speaker, because the writing must respect this constraint. In short, it does not agree with Latin.
In summary, you have to "invent" syllables that are pronounced in English as we would want to hear them in Latin.
"In the Word Builder you can choose from a list of Latin phrases, or you can type in your own (in English, phonetically, or in Votox) The program will convert English phrases to either, so if you don't get the results you want you can edit in a language that better controls the vowels and consonants."
EWQL Symphonic Choirs - Demo & Tutorial
You will see that there are those who have bumped into this, and they won't be long in making themselves heard
I have the previous version, not the PLAY... but the principles should be more or less the same...
T
Talentuoso
Member
Reply 4 by Talentuoso
Posts
206
I would also worry about the "colored panels" (i.e., the editor) that allow you to lengthen/shorten/mix individual phonemes; the color only represents the number of letters
If, for example, you care about a certain letter in a syllable, check that the horizontal reading in the TEXT EDITOR matches the vertical reading in the TIME EDITOR
You will see that it will improve significantly immediately.
B
Bit
Member
Reply 5 by Bit
Posts
225
lupusniger wrote:
1) why, while I am recording, can't I hear what the choirs are supposed to be singing? I can only do it by listening to what I recorded afterwards
Because the mechanism is "stupid" ... in the sense that it knows nothing about what is happening; it just passes signals on the MIDI channel at every note change.
No pitch change (in fact, pauses are ignored), no transition to the next syllable.
Which program are you using the plug-in with? Are you coming from programs like Finale or something like Cubase?
Why the mechanism is "stupid" ... in the sense that it knows nothing about what is happening; it just passes signals through the MIDI channel at every note change.
No pitch changes (pauses, in fact, are ignored), no transition to the next syllable.
I use Cubase 5 and this is the only plug-in that gives me this problem. If I use it as a stand-alone it doesn't do this... but in stand-alone mode I don't need it...
Talentuoso wrote:
I would also worry about the "colored panels" (i.e., the editor) that allow you to lengthen/shorten/mix individual phonemes; the color only represents the number of letters.
If, for example, you care about a certain letter in a syllable, check that the horizontal reading in the TEXT EDITOR matches the vertical reading in the TIME EDITOR.
You will see that it will improve significantly immediately.
eh??? I didn't quite understand this one... but I only know that the points where I have the most difficulty are the Ts at the end of words (which are practically not heard) and the word MEUS (which I write as two separate syllables, but the program "eats" the ME, resulting in all the other syllables being anticipated, preventing me from getting the result I want)...
...but all I know is that the points where I have the most difficulty are the Ts at the end of words (which are practically not heard) and the word MEUS (which I write as two separate syllables, but the program "eats" the ME, with the result of anticipating all the other syllables, preventing me from getting the result I want)...
The T at the end is objectively a problem because for English speakers it is silent; the only way to make them say it is to add a vowel... but usually this last one is unwanted.
Anyway, the T is a voiceless consonant [sic - user wrote 'vocale non sonora'], so even if it isn't heard in the general mix, I don't think it represents such an abysmal problem.
In any case, I would write it as "eST".
Did you know there are pre-packaged examples? Many words work the same way; try looking through those included in the product that work well.
As for the other problem, if it anticipates the other syllables and eats one up, in my opinion, it could depend on two things. I imagine you use the EW Symphonic Choir; do you know that, for example, syllables outside the range are skipped by the WB?
And the software's ranges, besides being absurd, do not coincide with real ones.
If I'm not mistaken, Alessandro Gozzo works specifically on your configuration; I will flag the topic to him.
it's been quite a while since I used it and it's so complicated that you forget immediately
anyway, as soon as I can, I'll try to look into it locally.
in the meantime, I'm telling you that you probably haven't set up Cubase correctly (since you can hear it in standalone mode).
tell me the steps you are taking.
as far as pronunciation goes, knowing how to use "the building blocks" is certainly important, but that is already an advanced level,
first of all, make sure you have selected the right voice, the Latin pronunciation and, above all, display the letters that it can understand (it doesn't actually read everything, but only part of the alphabet, and distinguishes between uppercase and lowercase). You will have to adapt the pronunciation to how it interprets those letters.
if you have the possibility, make a video showing the steps you take (both in opening the plugin on Cubase and in the writing), it would be very helpful to give you some tips.
As for the other problem, if it anticipates the other syllables and swallows one, in my opinion it could depend on two things. I imagine you are using EW Symphonic Choir; do you know that, for example, syllables outside the range are skipped by the WB?
And the software's extensions, besides being absurd, do not coincide with the real ones.
I know that.. but the first ME (the one almost at the beginning) pronounces it correctly... it is the MEUS one that doesn't pronounce it.
AleGozzo wrote:
Hi,
It's been quite a while since I used it and it's so complicated that you forget it immediately.
anyway, as soon as I can, I'll try some local troubleshooting.
in the meantime, I'll tell you that you probably haven't set up Cubase correctly (since you can hear it in standalone mode).
tell me the steps you are taking.
as far as pronunciation is concerned, it is certainly important to know how to use "the building blocks," but that is already an advanced level.
first of all, make sure you have selected the right voice, the Latin pronunciation and, above all, view the letters that it can understand (it doesn't actually read everything, but only part of the alphabet, and distinguishes between uppercase and lowercase). You will have to adapt the pronunciation to how it interprets those letters.
if you have the possibility, make a video showing the steps you take (both in opening the plugin in Cubase and in the writing), it would be very helpful to give you some tips.
So, I follow these steps:
1) I select the VST with F11
2) I choose EWSC and associate the MIDI track with it
3) I insert Wordbuilder as a MIDI insert
The problem is that in Cubase, connections between the various virtual cables cannot be made, as happens instead when I use it in standalone. If I use the Cubase virtual keyboard, you can see that the input arrives but no sound comes out. However, if I bypass the Wordbuilder, the sound can be heard...
ni il me mOl ce r!e PoTeST me us eST Se ne Fe ne do lo r!
But some things he pronounces as I would like, others not..
Example with "me"
In English, "me" is read as "mi", or am I wrong? See "Excuse me" to make ourselves understood.
Start from the assumption of making an Englishman say the word "me" by reading something, which cannot be exactly "me" ... at the cost of inventing it.
For me, what you wrote is very improvable; I don't have all the tools at hand right now, but the goal is to try, because it's not so obvious that by writing "me", the program will read "me" for you, in fact...
Already "meth"
Idem for the other cases.
Keep in mind that you won't be able to remove the English accent; you have to pretend that you have at your disposal an amateur choir of die-hard Englishmen :D
The problem is that in Cubase, the connections between the various virtual cables cannot be made, as happens instead when I use it in stand-alone mode. If I use the Cubase virtual keyboard, you can see that the input arrives but no sound comes out. However, if I bypass wordbuilder, the sound can be heard...
Well, for me it worked immediately without needing to mess around with cables.
I'll install it tonight and see; I hope to be able to tell you tomorrow.
I had made a clear diagram for Geppino who, if I recall correctly, had the same problem, but I can't find it on either edumus or mychance. If anyone remembers where it is, please let me know.
In English, "me" is read as "mi", or am I wrong? See "Excuse me" to make ourselves understood.
Start from the assumption of making an Englishman say the word "me" by reading something, which cannot be exactly "me" ... at the cost of inventing it.
For me, what you wrote is very improvable; I don't have all the tools at hand right now, but the goal is to try, because it's not so obvious that by writing "me", the program will read "me" to you, in fact...
Already "meth"
Idem for the other cases.
Keep in mind that you won't remove the English accent; you have to pretend that you have at your disposal an amateur choir of die-hard Englishmen :D
it reads "me" to me... it reads "mi" if I write mE.. go figure, these programmers...
AleGozzo wrote:
Well, it worked for me immediately without the need to mess around with cables.
I'll install it tonight and see; I hope to be able to tell you tomorrow.
I had made a clear diagram for Geppino who, if I recall correctly, had the same problem, but I can't find it on either edumus or mychance. If anyone remembers where it is, let me know.
I hope to hear from you... that diagram would be really useful to me.. but keep in mind that I have the Kompakt version, not the Play"}{
ni il me mOl ce r!e PoTeST me us eST Se ne Fe ne do lo r!
But some things he pronounces as I would like, others not..
Io would write it like this: nE El mOl C!e r!e po TeST me OS eST SE ne FE ne do lor!, exactly with this syllabification, otherwise everything gets a bit messed up. Then, to give a bit more realism, it is necessary that the syllables ending in a consonant are slightly detached from the beginning of the next syllable, because the WB pronounces the consonant at the end of the syllable when there is the release of the sound. If you play the phrase in real time, it is easy to perform this check; if you write the sequence of notes (as in Finale), every note will start exactly when the previous one concludes, without any breath between one note and the next, and then you will have an unnatural effect with the consonant of the previous syllable overlapping with that of the following syllable. In this case, you must act on the note durations and remove a fraction of duration to give the WB time to pronounce the preceding consonant without overlapping it with the next one. This could happen between the syllable El of "nihil" and mOL of "mulcere", between eST of "potest" and me of "meus", etc....
Answering the other questions: it is normal that you don't hear the WB in real time, because the WB is a Cubase insert, so it works on the output signal, not the input signal.
It can happen that some vowels are heard more loudly; it is a sampling defect. You can act on the velocities; anyway, to give more realism, you must act heavily on controllers cc11 (expression) and cc1 (modulation). The multi-samples used with the WB are of the dxf type (dynamic crossfade); it is advisable to use cc1 to obtain more realistic effects regarding dynamics.
If you have any other doubts about syllabification, I am at your disposal...
It can happen that some vowels are heard more loudly; it is a sampling defect, you can act on the velocities, however to provide more realism you must act heavily on the cc11 (expression) and cc1 (modulation) controllers. The multi-samples used with the wb are of the dxf (dynamic crossfade) type; it is advisable to use cc1 to obtain more realistic effects regarding dynamics.
If you have other doubts about syllabification, I am at your disposal...
Can you explain this part to me better? Generally, I work on audio files... it is the first time I am dealing with these commands..
These are MIDI commands, not audio, which serve to better control the various parameters of the sound you are using. Two very simple parameters are CC7 (volume) and CC10 (pan or stereophonic positioning, basically whether the sound is to the right, left, or center). Alongside these basic parameters, there are others that, in certain sounds from some libraries, can determine variations in the sound, such as a crescendo or a change in timbre. In EastWest libraries, the controller that regulates the instrument's volume is CC11 (expression), which is different from CC7 (which is the overall MIDI channel volume); in Garritan, it is CC1. The same CC1 (modulation) in EW regulates the dynamic crossfade in sounds that feature it and can determine changes in timbre or other highly effective effects; for example, in violins, it can regulate vibrato or attack, and in voices, the level of vibrato or sound intensity. These controls can be performed in real-time if you have a keyboard connected to the PC, or you can adjust them during the editing phase as I do using Finale.
You can find a list of the various MIDI controllers; they are the same for all libraries—they are GM, so to speak, shared by all brands—but in every VST brand, they can determine different effects.
One can intervene on the audio file during mastering, but in my opinion, it is better to intervene directly on the original sound to make the most of the samples' potential; in fact, if you read the Choir manual somewhere, it says to play a lot with CC1 to achieve greater realism, because no singer always sings at the same volume, much like instrumentalists. To achieve greater realism, some human "imperfections" are added. In Finale, there is also a command that allows for the randomization of various MIDI parameters specifically to make the performance less robotic; I think there is something similar in Cubase as well.
it seems strange to me that there isn't a way to hear it while playing the keyboard too, but it's a method I've never used.... have you tried tinkering in the PERIPHERALS / VST CONNECTIONS menu?
you had to act on the note durations and remove some fraction of duration to give the wb time to pronounce the preceding consonant without overlapping it with the next one. This could happen between the syllable El of "nihil" and mOL of "mulcere", between eST of "potest" and me of "meus", etc....
But the note durations I need to act on, is it the one in the midi editor (as I did) or the one in the wb?
