Non-verbal events2026-03-19·4 min read
MiMo-V2-TTS Non-Verbal Events
Pauses, breaths, sighs, coughs, and laughter often decide whether it sounds robotic or performed.
Rules (Avoid Overdoing It)
✓
Use events to support rhythm-don't add them to every line.
✓
Prioritize pauses: the most stable and least intrusive enhancement.
✓
Breaths/coughs fit nervous/running/weak scenes; soft laughs fit comedy/sarcasm; sighs and long pauses fit sadness/tension.
Copyable Examples
Short pause
I... didn't mean that. (short pause) I just didn't know how to say it.
Long pause + sigh
If I had... (silence) (long sigh) would it have been different?
Breath + nervous
(deep breath) Calm down. It's just an interview... I can do this.
Cough + weak
Water... please... (coughing) I can't stop-my throat is burning.
Soft laugh + contrast
That's ridiculous... (soft laugh) Fine, I won't argue with you.