# Ranting/Angry Style — VoiceNet dimension guide

*Category:* Speaking Style · *code:* `vn_S_RANT` · *slot:* 94

This dimension measures how much the voice sounds like an angry rant. Merge the `__high` / `__low` LoRA from `laion/moss-voicenet-dimension-loras` at 25/50/75/100/125%. Costs are shifts vs. the 24-sample neutral baseline.

**TL;DR.** **Ranting/Angry Style** controls how much the voice sounds like an angry rant. The **__high** LoRA shifts it +0.75 at 125% (worst cost: burst-blend -0.40); the **__low** LoRA shifts it -0.02 at 25% (worst cost: genuineness -0.02). The **__high** direction is the stronger, more reliable knob here.

## ▲ HIGH — Ranting/Angry Style __high

| dose | target | Δtarget | Δgenu | Δblend | Δqual |
|---|---|---|---|---|---|
| baseline (0%) | 0.024 | — | — | — | — |
| 25% | 0.047 | +0.02 | -0.02 | -0.08 | -0.10 |
| 50% | 0.275 | +0.25 | -0.03 | -0.28 | -0.09 |
| 75% | 0.462 | +0.44 | -0.05 | -0.37 | -0.20 |
| 100% | 0.660 | +0.64 | -0.05 | -0.43 | -0.29 |
| 125% ★ | 0.770 | +0.75 | -0.07 | -0.40 | -0.32 |

- **EmoNet emotion** — ↑ Triumph (+0.87), Anger (+0.74), Pride (+0.66); ↓ Relief (-0.68), Affection (-0.67), Contentment (-0.65)
- **VoiceNet dimension** — ↑ Attack (+0.94), Dynamic Arc (+0.93), Arousal (+0.92); ↓ Monologue Style (-0.88), Whisper-Talk Style (-0.79), Warmth (-0.79)

Merging **Ranting/Angry Style __high** drives the dimension: at the recommended **125%** dose the measured Ranting/Angry Style score moves from a 0.02 baseline to 0.77 (a +0.75 shift). The dose-response is clean and monotonic — each step pushes the dimension further. That reach is a steep cost — the biggest hit is to **burst-blend** (-0.40); across the board genuineness -0.07, burst-blend -0.40, speech quality -0.32. Emotionally it reads as Triumph (r 0.87), Anger (0.74) and Pride (0.66), and it moves away from Relief, Affection and Contentment. On other voice dimensions it co-moves most with Attack (0.94), Dynamic Arc (0.93) and Arousal (0.92), while pulling down Monologue Style, Whisper-Talk Style and Warmth. Practical call: merge __high at **100–125%**; expect Triumph and Attack to rise and burst-blend to fall. If you don’t want Triumph bleeding in, offset it with a prompt cue toward Relief to keep the ranting/angry style shift without the side-effect.

## ▼ LOW — Ranting/Angry Style __low

| dose | target | Δtarget | Δgenu | Δblend | Δqual |
|---|---|---|---|---|---|
| baseline (0%) | 0.024 | — | — | — | — |
| 25% ★ | 0.000 | -0.02 | -0.02 | +0.01 | +0.02 |
| 50% | 0.000 | -0.02 | -0.03 | -0.07 | +0.07 |
| 75% | 0.006 | -0.02 | -0.08 | -0.14 | +0.09 |
| 100% | 0.000 | -0.02 | +0.04 | -0.01 | +0.03 |
| 125% | 0.000 | -0.02 | -0.04 | -0.01 | +0.11 |

- **EmoNet emotion** — ↑ Doubt (+0.67), Jealousy & Envy (+0.43), Fear (+0.34); ↓ Contemplation (-0.28), Emotional Numbness (-0.21), Sexual Lust (-0.20)
- **VoiceNet dimension** — ↑ Tension (+0.54), Volatility (+0.51), Dramatic Style (+0.44); ↓ Warmth (-0.53), Esthetics (-0.47), Recording Quality (-0.45)

Merging **Ranting/Angry Style __low** pulls down the dimension: at the recommended **25%** dose the measured Ranting/Angry Style score moves from a 0.02 baseline to 0.00 (a -0.02 shift). The dose-response is monotonic but it saturates early, so higher doses add little. That reach is nearly free — the biggest hit is to **genuineness** (-0.02); across the board genuineness -0.02, burst-blend +0.01, speech quality +0.02. Emotionally it reads as Doubt (r 0.67), Jealousy & Envy (0.43) and Fear (0.34), and it moves away from Contemplation, Emotional Numbness and Sexual Lust. On other voice dimensions it co-moves most with Tension (0.54), Volatility (0.51) and Dramatic Style (0.44), while pulling down Warmth, Esthetics and Recording Quality. Practical call: merge __low at **a light 25–50%**; expect Doubt and Tension to rise and genuineness to fall. If you don’t want Doubt bleeding in, offset it with a prompt cue toward Contemplation to keep the ranting/angry style shift without the side-effect.

---
[← all dimensions](index.md) · [human/audio version](../../site/dims/S_RANT.html)