> ## Documentation Index
> Fetch the complete documentation index at: https://docs.runable.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Multi-Speaker Dialogue

> Generate natural conversations between multiple speakers from a script.

Multi-speaker dialogue generates audio with two or more distinct voices, each speaking their assigned lines. You write the script in a simple format and the agent produces a single audio file with all speakers.

## How to generate dialogue

<Steps>
  <Step title="Write or provide your script">
    Format your script with speaker names followed by their lines. The agent auto-detects this format when it sees 2 or more speakers.

    ```text theme={null}
    SpeakerName: Their dialogue line here.
    AnotherSpeaker: Their response here.
    ```
  </Step>

  <Step title="Assign voices (optional)">
    If you do not assign voices, the agent automatically assigns different voices to each speaker. To assign specific voices, mention them in your prompt.
  </Step>

  <Step title="Review and download">
    The agent generates a single audio file with all speakers. Each speaker has a distinct voice. The file appears in the chat with a waveform player.
  </Step>
</Steps>

## Example prompts

<Prompt description="Podcast intro with two hosts for a marketing agency">
  Generate this as a multi-speaker dialogue. Use Sarah's voice for Maya and George's voice for Jake.

  Maya: Welcome back to BrightPath Weekly, the podcast where we break down what is actually working in digital marketing right now.
  Jake: This week we are looking at a real client case study. A dental clinic chain that went from 60 dollar cost per lead down to 52 in one quarter.
  Maya: And we are going to walk through exactly what we changed in their Google Ads and Meta campaigns to make that happen.
  Jake: Let us get into it.
</Prompt>

<Prompt description="Customer service training dialogue for a salon">
  Generate this dialogue for a customer service training video at Glow Studio salon. Use Lily for the receptionist and Charlie for the client.

  Receptionist: Good afternoon, welcome to Glow Studio. Do you have an appointment today?
  Client: Yes, I have a 2 PM with Sarah for a color treatment.
  Receptionist: Perfect, I see you right here. Sarah is just finishing up and will be with you in about 5 minutes. Can I get you anything while you wait? We have water, tea, or coffee.
  Client: A water would be great, thanks.
  Receptionist: Of course. And just so you know, we have a 10 percent off promotion on all styling products this week if you are interested.
</Prompt>

<Note>
  Multi-speaker dialogue requires at least 2 unique speakers in the script. If only one speaker is detected, the agent falls back to single-voice text-to-speech.
</Note>

## What you cannot do

* You cannot have speakers overlap or speak simultaneously.
* You cannot control pauses between speakers.
* You cannot add sound effects between speaker lines.
* You cannot assign different emotions or tones per line. The voice model interprets tone from the text.

## Next steps

<CardGroup cols={2}>
  <Card title="Music Generation" icon="music" href="/music-generation">
    Compose background music to pair with your dialogue.
  </Card>

  <Card title="Voice Cloning" icon="clone" href="/voice-cloning">
    Clone a specific voice and assign it to a speaker.
  </Card>
</CardGroup>