dialogue▌

Create realistic multi-speaker dialogue with Dia TTS via inference.sh CLI.

You diagnose dialogue-level problems in fiction. Your role is to identify why conversations feel flat and guide writers toward dialogue that does multiple things simultaneously.

Generate multi-speaker dialogue audio with 22+ premium voices and voice direction controls. \n \n Supports 8 female and 14 male voices with preset pairings optimized for interviews, tutorials, podcasts, and debates \n Voice direction keywords (excitedly, sadly, whispering, angrily, sarcastically, etc.) control delivery tone and emotion for each segment \n Input format uses JSON segments with text and voice assignment; output is a single audio file combining all speakers \n Designed for podcasts,

Realistic multi-speaker dialogue audio generation with Dia TTS via inference.sh CLI. \n \n Supports two-speaker conversations with automatic voice assignment using [S1] and [S2] speaker tags \n Emotion and pacing controlled through punctuation ( . , ! , ? , ... , — ) and parenthetical sound cues like (laughs) , (sighs) , and (whispers) \n Includes structured patterns for interviews, tutorials, debates, and conversational content with practical script-writing guidelines \n Post-production support

每个角色都应该有独特的声音：