← all dimensions · markdown (agent skill) · manual home
Mask Resonance VoiceNet dimension guide
Resonance Placementcode vn_R_MASKslot 78
This dimension measures resonance from the mask / sinus region. Below: the merge-scaling dose-response for the __high and __low LoRA (laion/moss-voicenet-dimension-loras) at 25/50/75/100/125%, the measured target value at each dose, the top-3 ± correlated EmoNet emotions and VoiceNet dimensions, three audio samples per dose, and a grounded prompting tip. Costs are shifts vs. the 24-sample neutral baseline in genuineness, vocal-burst blend and speech quality.
Mask Resonance controls resonance from the mask / sinus region. The __high LoRA shifts it +0.39 at 125% (worst cost: burst-blend -0.24); the __low LoRA shifts it -0.09 at 50% (worst cost: burst-blend -0.15). The __high direction is the stronger, more reliable knob here.
▲ HIGH — Mask Resonance __high
Dose-response
| dose | target | Δtarget | Δgenu | Δblend | Δqual |
|---|
| baseline (0%) | 0.123 | — | — | — | — |
| 25% | 0.102 | -0.02 | -0.03 | -0.04 | -0.01 |
| 50% | 0.268 | +0.14 | -0.05 | -0.20 | +0.04 |
| 75% | 0.317 | +0.19 | -0.05 | -0.17 | +0.06 |
| 100% | 0.449 | +0.33 | -0.07 | -0.25 | +0.08 |
| 125% ★ | 0.513 | +0.39 | -0.01 | -0.24 | +0.08 |
EmoNet emotion correlation
Concentration+0.42
Embarrassment+0.18
Thankfulness/Gratitude+0.10
Longing-0.59
Infatuation-0.50
Fear-0.49
VoiceNet dimension correlation
Nasal Resonance+0.92
Brightness+0.90
Head Resonance+0.88
Perceived Gender-0.77
Cognitive Load-0.67
Roughness-0.61
Listen — 3 samples per dose
25%measured target: 0.05 · 0.13 · 0.00
50%measured target: 0.40 · 0.14 · 0.56
75%measured target: 0.41 · 0.65 · 0.28
100%measured target: 0.76 · 0.15 · 0.44
125% ★measured target: 0.68 · 0.42 · 0.44
Merging Mask Resonance __high raises the dimension: at the recommended 125% dose the measured Mask Resonance score moves from a 0.12 baseline to 0.51 (a +0.39 shift). The dose-response is clean and monotonic — each step pushes the dimension further. That reach is a real cost — the biggest hit is to burst-blend (-0.24); across the board genuineness -0.01, burst-blend -0.24, speech quality +0.08. Emotionally it reads as Concentration (r 0.42), Embarrassment (0.18) and Thankfulness/Gratitude (0.10), and it moves away from Longing, Infatuation and Fear. On other voice dimensions it co-moves most with Nasal Resonance (0.92), Brightness (0.90) and Head Resonance (0.88), while pulling down Perceived Gender, Cognitive Load and Roughness. Practical call: merge __high at 100–125%; expect Concentration and Nasal Resonance to rise and burst-blend to fall. If you don’t want Concentration bleeding in, offset it with a prompt cue toward Longing to keep the mask resonance shift without the side-effect.
▼ LOW — Mask Resonance __low
Dose-response
| dose | target | Δtarget | Δgenu | Δblend | Δqual |
|---|
| baseline (0%) | 0.123 | — | — | — | — |
| 25% | 0.094 | -0.03 | +0.00 | -0.05 | -0.05 |
| 50% ★ | 0.030 | -0.09 | -0.01 | -0.15 | -0.12 |
| 75% | 0.061 | -0.06 | -0.06 | -0.08 | -0.04 |
| 100% | 0.070 | -0.05 | -0.01 | -0.10 | +0.01 |
| 125% | 0.069 | -0.05 | -0.03 | -0.10 | -0.04 |
EmoNet emotion correlation
Contentment+0.38
Concentration+0.33
Embarrassment+0.23
Sexual Lust-0.33
Infatuation-0.29
Anger-0.28
VoiceNet dimension correlation
Head Resonance+0.79
Register+0.75
Nasal Resonance+0.67
Perceived Gender-0.53
Roughness-0.47
Disfluency-0.46
Listen — 3 samples per dose
25%measured target: 0.27 · 0.00 · 0.39
50% ★measured target: 0.08 · 0.00 · 0.00
75%measured target: 0.00 · 0.12 · 0.00
100%measured target: 0.17 · 0.00 · 0.06
125%measured target: 0.00 · 0.12 · 0.05
Merging Mask Resonance __low pulls down the dimension: at the recommended 50% dose the measured Mask Resonance score moves from a 0.12 baseline to 0.03 (a -0.09 shift). The dose-response is mostly ordered but noisy between doses. That reach is a modest cost — the biggest hit is to burst-blend (-0.15); across the board genuineness -0.01, burst-blend -0.15, speech quality -0.12. Emotionally it reads as Contentment (r 0.38), Concentration (0.33) and Embarrassment (0.23), and it moves away from Sexual Lust, Infatuation and Anger. On other voice dimensions it co-moves most with Head Resonance (0.79), Register (0.75) and Nasal Resonance (0.67), while pulling down Perceived Gender, Roughness and Disfluency. Practical call: merge __low at 25–75%; expect Contentment and Head Resonance to rise and burst-blend to fall. If you don’t want Contentment bleeding in, offset it with a prompt cue toward Sexual Lust to keep the mask resonance shift without the side-effect.