Updated flag list

Tutorial Updated flag list 1.1

susrever

Ruko's Ruffians
Defender of Defoko
susrever submitted a new resource:

Updated flag list - list of all the knonw utau flags and a brief description of them

I found extremely frustrating having to search old topics to guess what a flag might do,and if my resampler of choice supported it, so I decided to make a flag compatibility table.

It includes a table with all the currently avaible resamplers and the flag they support and a list of the flags and their descriptions.

I have checked the readmes of all the resamplers that came with one, I've tested most of them, and I've checked for errors twice, but errors are bound to be made. Please report...

Read more about this resource...
 

Kiyoteru

UtaForum power user
Supporter
Defender of Defoko
I think it'd be handy to have information on the resource page itself! Having a downloadable PDF seems like a bonus.

Oh, and here's a full list of Moresampler's flags. I think a few of the Moresampler-specific flags might be missing from your resource, and the behavior of "b" is actually different here than in other resamplers (namely, it actually becomes useful)
g (-100, 100) (default 0)

Alter the perceived gender of the voice. Positive: male; negative: female.

t (-1200, 1200) (default 0)

Shift the pitch by certain cents. 1 cent = 1/100 semitone.

P (0, 100) (default 86)

Peak compressor. When set to 100, it normalizes the peak of output waveform to half the maximum level. When set to 0, it doesn't normalize the output at all. For a number between 0 and 100, the degree of normalization is interpolated. This flag is only effective when "synthesis-utau-style-normalization" is not off.

A (-100, 100) (default 0)

Amplitude modulation. This flags modulates the amplitude in correlation with change of pitch. It could be helpful for creating realistic vibratos. The sign (positive or negative) controls the direction of such modulation.

b (-20, 100) (default 0)

Amplitude gain for unvoiced consonants. This flag amplifies or attenuates unvoiced consonants (e.g. /t/ /k/ /s/) by a factor of 0.05 times the number after b. It has less or no effect for voiced consonants (e.g. /g/, /m/). E

e (no range) (no default)

Force Moresampler to extend sustained vowels by stretching (time scaling), as opposed to looping. A related flag is Me, which has the exactly opposite effect (force looping). By default the duration extension method is specified by "synthesis-duration-extension-method" option.

Mt (-100, 100) (default 0)

Tenseness - the extent to which the vocal folds are stressed or relaxed. Positive values correspond to tenser voice quality and vice versa.

Mb (-100, 100) (default 0)

Breathiness. Positive values correspond to breathier voice and negative values reduce the breathing noise. When set to 100, the voice completely becomes whispering.

Mo (-100, 100) (default 0)

Openness - the degree of jaw opening during phonation. Positive values correspond to wide opening and vice versa.

Md (-100, 100) (default 0)

Dryness - the degree of amplitude modulation received by breathing noise due to the periodicity of glottal air flow. The effect of this flag is very subtle and mostly takes place in high frequency band (usually above 6kHz).

Ms (0, 10) (default 0)

Stabilization - fixing the occasional 'pops' that mostly occurs when shifting down the pitch. It is basically a runtime version of "analysis-anti-distortion" with adjustable degree of stabilization (higher number corresponds to stronger stabilization). This flag is recommended when the popping only occasionally occurs, otherwise turn on analysis-anti-distortion instead.

Mm (0, 100) (default 100)

Model interpolation - interpolating between the classical speech model used before version 0.3.0 and the novel model used since then. By default Moresampler uses the new model (Mm100).

ME (-100, 100) (default 0)

Formant emphasis - given positive values, it emphasizes the formants; given negative values, the voice becomes fuzzy.

Me (no range) (no default)

Force looping- the opposite of 'e' flag.

MC (0, 100) (default 0)

Coarseness. It replaces voice by a harsh roaring similar to death growl.

MG (0, 100) (default 0)

Growl: shivering, powerful voice effect.

MD (0, 100) (default 0)

Distortion: noisy, sharp, and a bit metallic voice effect.

Mr (-100, 100) (default 0)

Creates a "singer's formant" around 3kHz.

Mp (0, 100) (default 0)
Adds a random perturbation to pitch curve

---

Another handy feature added is called "meta flags". While you may have found combining several flags (e.g. "MG50MD30MC20Mb30Mt50") to be useful, typing those alphabets and numbers on each note could be somewhat time-wasting. Now with meta flags you can shorten a flag sequence to a single flag. For example, having the following line in moreconfig.txt defines "M1" to be "MG50MD30MC20Mb30Mt50",

meta-flag-1 MG50MD30MC20Mb30Mt50

When you refer to "M1" in UTAU's note settings, Moresampler replaces "M1" by "MG50MD30MC20Mb30Mt50" while keeping other flags intact. Well, the case is actually more complicated when there's overlap between meta flags and ordinary flags or other meta flags, but the general rule is that what comes first will always overwrite what comes later, as illustrated in the following examples,

eMG20M1 -> eMG20MD30MC20Mb30Mt50

eM1MG20 -> eMG50MD30MC20Mb30Mt50

eMb25MG30M1Mt60 -> eMG30MD30MC20Mb25Mt50

By adding dot and a number after a meta flag, you can scale the underlying flags by a certain percentage. For instance "M1.50" evaluates to "MG25MD15MC10Mb15Mt25". The scaling factor can even be negative. For flags without a parameter, the scaling has no effect, even when the number is zero. A slight difference regarding scaling and overwriting rules is that when scaled meta flags share a set of overlapping flags, those flags will be "blended" together, instead of being overwritten. If you define "M2" as "Mb-50MG50", then "M1.50M2.-50" would be the same as "MG0MD15MC10Mb40Mt25".
 
Last edited:
  • Like
Reactions: partial

susrever

Ruko's Ruffians
Defender of Defoko
Thread starter
susrever updated Updated flag list with a new update entry:

updated flag list : added a non pdf-version, updated the pdf version, corrected some mistakes.

As @Kiyoteru suggested, here it is the guide, avaible in the post so that people don't have to download the pdf!
Also, updated some flags and added others that I forgot to add, thanks again @Kiyoteru. (Namely, the flags I forgot were Mr and Mp, and the descriptions for b and A have been updated.)
The pdf also has been updated, so if you had downloaded it already, I suggest you download the updated version!

· $direct = true
Range :...

Read the rest of this update entry...
[doublepost=1538641487][/doublepost]Thank you for your suggestions! I have made the guide avaible also in a non-pdf way, so it's easier for people to use this resource, and have added a few more moresamplers flags I forgot.
I have not added t, p, and g because they are pretty much standard flags and they work the same for the other resamplers, but I have indeed updated A and b ! Thanks again for spotting those mistakes.
 

lunari162

Defoko's Slaves
Defender of Defoko
This is a really useful resource, so thank you for creating it! I think there might be one small error, though:
· g
Range : -99 / 99
Default : 0
Effect : Values from -99 to 0 make the voice sound more masculine
Values from 0 to 99 make the voice sound more feminine.
I'm pretty sure the masculine/feminine part is reversed; negative values (-99 to 0) make the voice more feminine, while positive ones (0 to 99) make it more masculine. I might be reading it wrong, but I figured it couldn't hurt to bring it up. If I did end up reading it incorrectly, just ignore this ^^;
 

susrever

Ruko's Ruffians
Defender of Defoko
Thread starter
I'm pretty sure the masculine/feminine part is reversed; negative values (-99 to 0) make the voice more feminine, while positive ones (0 to 99) make it more masculine. I might be reading it wrong, but I figured it couldn't hurt to bring it up. If I did end up reading it incorrectly, just ignore this ^^;
asdfghjkl me fail, thank you for spotting it ! Will correct post ASAP and update pdf when I get home ;A;;
 

Kiyoteru

UtaForum power user
Supporter
Defender of Defoko
Please update the information for the W flag as supported by tn_fnds.

W-1: Disable pitch shifting
W(any other value): Specify original pitch in Hz, ignoring the pitch analyzed from the audio. For example, W440 will make tn_fnds act as if the original sample was recorded at A4.
 

Similar threads