← all dimensions · markdown (agent skill) · manual home
Brightness VoiceNet dimension guide
Spectral & Timbralcode vn_BRGTslot 61
This dimension measures the prominence of high-frequency energy in the voice. 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.
Brightness controls the prominence of high-frequency energy in the voice. The __high LoRA shifts it +0.30 at 100% (worst cost: burst-blend -0.20); the __low LoRA shifts it -0.04 at 50% (worst cost: speech quality -0.07). The __high direction is the stronger, more reliable knob here.
▲ HIGH — Brightness __high
Dose-response
| dose | target | Δtarget | Δgenu | Δblend | Δqual |
|---|
| baseline (0%) | 0.177 | — | — | — | — |
| 25% | 0.197 | +0.02 | +0.04 | -0.13 | -0.13 |
| 50% | 0.238 | +0.06 | -0.06 | -0.12 | +0.06 |
| 75% | 0.283 | +0.11 | -0.04 | -0.14 | -0.03 |
| 100% ★ | 0.479 | +0.30 | -0.05 | -0.20 | +0.06 |
| 125% | 0.416 | +0.24 | -0.05 | -0.33 | -0.07 |
EmoNet emotion correlation
Hope/Enthusiasm/Optimism+0.44
Elation+0.44
Pleasure/Ecstasy+0.43
Longing-0.42
Helplessness-0.40
Fear-0.40
VoiceNet dimension correlation
Arousal+0.85
Mask Resonance+0.84
Chunking+0.83
Perceived Gender-0.65
Cognitive Load-0.63
Vulnerability-0.52
Listen — 3 samples per dose
25%measured target: 0.03 · 0.25 · 0.33
50%measured target: 0.14 · 0.50 · 0.16
75%measured target: 0.26 · 0.27 · 0.21
100% ★measured target: 0.56 · 0.48 · 0.53
125%measured target: 0.18 · 0.08 · 0.65
Merging Brightness __high drives the dimension: at the recommended 100% dose the measured Brightness score moves from a 0.18 baseline to 0.48 (a +0.30 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 (-0.20); across the board genuineness -0.05, burst-blend -0.20, speech quality +0.06. Emotionally it reads as Hope/Enthusiasm/Optimism (r 0.44), Elation (0.44) and Pleasure/Ecstasy (0.43), and it moves away from Longing, Helplessness and Fear. On other voice dimensions it co-moves most with Arousal (0.85), Mask Resonance (0.84) and Chunking (0.83), while pulling down Perceived Gender, Cognitive Load and Vulnerability. Practical call: merge __high at 75–100%; expect Hope/Enthusiasm/Optimism and Arousal to rise and burst-blend to fall. If you don’t want Hope/Enthusiasm/Optimism bleeding in, offset it with a prompt cue toward Longing to keep the brightness shift without the side-effect.
▼ LOW — Brightness __low
Dose-response
| dose | target | Δtarget | Δgenu | Δblend | Δqual |
|---|
| baseline (0%) | 0.177 | — | — | — | — |
| 25% | 0.161 | -0.02 | -0.02 | -0.11 | -0.01 |
| 50% ★ | 0.140 | -0.04 | +0.02 | -0.06 | -0.07 |
| 75% | 0.199 | +0.02 | -0.06 | -0.13 | +0.04 |
| 100% | 0.142 | -0.04 | -0.05 | -0.17 | -0.07 |
| 125% | 0.162 | -0.01 | -0.02 | -0.10 | -0.11 |
EmoNet emotion correlation
Amusement+0.31
Contentment+0.27
Hope/Enthusiasm/Optimism+0.20
Fear-0.28
Anger-0.28
Pride-0.23
VoiceNet dimension correlation
Head Resonance+0.66
Mask Resonance+0.66
Register+0.57
Perceived Gender-0.47
Disfluency-0.40
Cognitive Load-0.40
Listen — 3 samples per dose
25%measured target: 0.16 · 0.17 · 0.15
50% ★measured target: 0.11 · 0.09 · 0.18
75%measured target: 0.52 · 0.29 · 0.03
100%measured target: 0.35 · 0.10 · 0.22
125%measured target: 0.21 · 0.30 · 0.12
Merging Brightness __low dials back the dimension: at the recommended 50% dose the measured Brightness score moves from a 0.18 baseline to 0.14 (a -0.04 shift). The dose-response is mostly ordered but noisy between doses. That reach is a modest cost — the biggest hit is to speech quality (-0.07); across the board genuineness +0.02, burst-blend -0.06, speech quality -0.07. Emotionally it reads as Amusement (r 0.31), Contentment (0.27) and Hope/Enthusiasm/Optimism (0.20), and it moves away from Fear, Anger and Pride. On other voice dimensions it co-moves most with Head Resonance (0.66), Mask Resonance (0.66) and Register (0.57), while pulling down Perceived Gender, Disfluency and Cognitive Load. Practical call: merge __low at 25–75%; expect Amusement and Head Resonance to rise and speech quality to fall. If you don’t want Amusement bleeding in, offset it with a prompt cue toward Fear to keep the brightness shift without the side-effect.