Create multi-speaker dialogue with ElevenLabs Text to Dialogue. Hear a real audio drama, map voices to lines, review pricing, and copy API code.
Provider: ElevenLabs
Category: music generation
Endpoint: /v1/tasks
Status: Available
Cost: See live page pricing
Create multi-speaker dialogue with ElevenLabs Text to Dialogue. Hear a real audio drama, map voices to lines, review pricing, and copy API code.
Pricing
Standard Usage
Write the conversation line by line; each line can use a different voice. Total characters across all lines must not exceed 5000; At least 1 items
0 = more emotional and varied, 0.5 = balanced, 1 = most stable and consistent
Auto-detects the language by default; you can also set it manually
Ready for generation
Configure your parameters and click "Run" to see the output here.
Review the script, speaker mapping, and generated audio together to judge speaker separation and dialogue flow.
Three speakers alternate across a short suspense scene with a stable voice assigned to every line.
Dialogue script
Brian: Midnight. The city hummed beneath a curtain of rain. Laura: You're late. I almost gave up on you. Mark: The bridge was watched. I took the long way. Laura: Then we don't have much time.
Choose an adjacent route for music, single-speaker speech, or multi-speaker dialogue.
High-quality narration
Higher-quality voiceovers for video narration and spoken content.
Create multi-speaker dialogue with ElevenLabs Text to Dialogue. Hear a real audio drama, map voices to lines, review pricing, and copy API code.
The Playground, listening examples, and API Quickstart use the same model ID: elevenlabs/text-to-dialogue.
These specifications reflect the current ElevenLabs Text to Dialogue request contract.
Use three steps to organize the input, validate the audio, and integrate the API.
Put each line in a separate dialogue item with clear speaker roles.
Reuse the same voice for each character, then set language and stability when needed.
Check speaker separation, pauses, and dialogue flow before production integration.
Copy a minimal request. See Docs for complete parameters, task retrieval, and errors.
POST /v1/taskselevenlabs/text-to-dialogueThe create request immediately returns a taskId; audio generation continues asynchronously.
Dialogue is an array of lines, each with text and a voice. Reuse the same voice for a character to keep that speaker consistent.
The combined dialogue.text content can contain up to 5,000 characters. Language can be detected automatically or set with language_code.
HiAPI bills effective dialogue text characters and displays the rate per 1,000 characters. See the live page price and task record for the final charge.