Voice Persona

This page covers voice prompt design principles, neural voice styling, and SSML markup control.

Conversational Voice Prompt Engineering

Content coming soon.

Designing system prompts for natural spoken dialogue
Eliminating visual artifacts (markdown tables, bullet points, raw URLs)
Phonetic formatting for acronyms, numbers, and currency values

SSML Markup & Prosody Control

Content coming soon.

Pause insertion (<break time="250ms"/>)
Speech rate and pitch modulation (<prosody>)
Pronunciation lexicons and phonetic spellings (<phoneme>)

Latency-Masking Utterances (Filler Audio)

Content coming soon.

Dynamic filler generation during background database queries and tool calls
Contextual fillers (“Let me check that order for you…”)