AI Voice Actors: Turn a Dialogue Script into Multi-Speaker Audio

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

ElevenLabs

AI Voice Actors: Turn a Dialogue Script into Multi-Speaker Audio

Give each line of your dialogue its own voice and get one continuous multi-speaker MP3 from ElevenLabs eleven_v3 — up to 20 lines and 10 voices, with delivery tags. You bring the script and voices; no cloning, no script writing. Pay per use via Toolmu, no subscription, max $100/run.

You receive

Returns one continuous MP3 of your dialogue, performed in order by the ElevenLabs voices you assigned — up to 20 lines and 10 distinct voices per run; eleven_v3 tags shape the delivery. Audio only: no transcript, no per-line files, no music, no video, and no script written from a topic.

Start with your task

Use multiple AI voice actors to turn [Mia: Hi, Ben! / Ben: Hello, Mia!] into one continuous conversation. Choose [a voice name from the available list] for each speaker. Provide your own lines; the tool does not write a script for you.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
body`body`	`object`	Required	Convert a multi-speaker dialogue to speech audio Required fields `inputs`
inputs`body.inputs`	`array`	Required	The dialogue: ordered text + voice_id pairs, rendered as one continuous audio file. Minimum items `1` Maximum items `20`
[]`body.inputs[]`	`object`	—	Required fields `text, voice_id`
text`body.inputs[].text`	`string`	—	The line to speak. Supports eleven_v3 audio tags like [giggling], [curious], [whispering]. Minimum length `1`
voice_id`body.inputs[].voice_id`	`string`	—	The ElevenLabs voice for this line — pick voice_ids from /voices (max 10 unique voices per request). Minimum length `1`
model_id`body.model_id`	`string`	Optional	Dialogue model (eleven_v3 — most expressive). Default `eleven_v3` Options `eleven_v3`
settings`body.settings`	`object`	Optional	Optional generation settings, passed through verbatim.
stability`body.settings.stability`	`number`	Optional	Voice stability (0–1). Lower = broader emotional range. Minimum `0` Maximum `1`
language_code`body.language_code`	`string`	Optional	Optional ISO 639-1 language code to enforce. Minimum length `2` Maximum length `8`

View raw JSON

{
  "type": "object",
  "required": [
    "body"
  ],
  "properties": {
    "body": {
      "type": "object",
      "$schema": "https://json-schema.org/draft/2020-12/schema",
      "required": [
        "inputs"
      ],
      "~standard": {
        "vendor": "zod",
        "version": 1,
        "jsonSchema": {}
      },
      "properties": {
        "inputs": {
          "type": "array",
          "items": {
            "type": "object",
            "required": [
              "text",
              "voice_id"
            ],
            "properties": {
              "text": {
                "type": "string",
                "minLength": 1,
                "description": "The line to speak. Supports eleven_v3 audio tags like [giggling], [curious], [whispering]."
              },
              "voice_id": {
                "type": "string",
                "minLength": 1,
                "description": "The ElevenLabs voice for this line — pick voice_ids from /voices (max 10 unique voices per request)."
              }
            },
            "additionalProperties": false
          },
          "maxItems": 20,
          "minItems": 1,
          "description": "The dialogue: ordered text + voice_id pairs, rendered as one continuous audio file."
        },
        "model_id": {
          "enum": [
            "eleven_v3"
          ],
          "type": "string",
          "default": "eleven_v3",
          "description": "Dialogue model (eleven_v3 — most expressive)."
        },
        "settings": {
          "type": "object",
          "properties": {
            "stability": {
              "type": "number",
              "maximum": 1,
              "minimum": 0,
              "description": "Voice stability (0–1). Lower = broader emotional range."
            }
          },
          "description": "Optional generation settings, passed through verbatim.",
          "additionalProperties": false
        },
        "language_code": {
          "type": "string",
          "maxLength": 8,
          "minLength": 2,
          "description": "Optional ISO 639-1 language code to enforce."
        }
      },
      "description": "Convert a multi-speaker dialogue to speech audio",
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}

Result schema

Output fields have not been documented yet.

Make a short two-person conversation
Turn [Mia: Hi! / Ben: Hello!] into one recording, choosing [a voice name] for Mia and [another voice name] for Ben.
Get connected
Give three characters different voices
Turn [a dialogue between three characters] into one conversation and choose [a voice name] for each person.
Use multiple AI voice actors to turn [Mia: Hi, Ben! / Ben: Hello, Mia!] into one continuous conversation. Choose [a voice name from the available list] for each speaker. Provide your own lines; the tool does not write a script for you.
Use a script you already wrote
Use [your dialogue script] and [the voice names you want for each speaker] to make one continuous recording.

You receive

Start with your task

Inputs & outputs

Input schema

Result schema

Make a short two-person conversation

Give three characters different voices

Use a script you already wrote