Turn text into speech with ElevenLabs

{
  "type": "object",
  "required": [
    "body"
  ],
  "properties": {
    "body": {
      "type": "object",
      "$schema": "https://json-schema.org/draft/2020-12/schema",
      "required": [
        "text"
      ],
      "~standard": {
        "vendor": "zod",
        "version": 1,
        "jsonSchema": {}
      },
      "properties": {
        "text": {
          "type": "string",
          "examples": [
            "Hello! Your order has shipped and arrives Thursday."
          ],
          "maxLength": 5000,
          "minLength": 1,
          "description": "A string of 1–5,000 characters to read aloud. Split longer text into separate runs."
        },
        "model_id": {
          "enum": [
            "eleven_multilingual_v2",
            "eleven_flash_v2_5",
            "eleven_v3"
          ],
          "type": "string",
          "default": "eleven_multilingual_v2",
          "description": "Choose eleven_multilingual_v2, eleven_flash_v2_5 or eleven_v3. Defaults to eleven_multilingual_v2."
        },
        "voice_id": {
          "type": "string",
          "default": "21m00Tcm4TlvDq8ikWAM",
          "minLength": 1,
          "description": "A nonempty ElevenLabs voice ID. Defaults to 21m00Tcm4TlvDq8ikWAM. A custom value must identify a valid, accessible voice."
        },
        "voice_settings": {
          "type": "object",
          "properties": {
            "speed": {
              "type": "number",
              "maximum": 1.2,
              "minimum": 0.7,
              "description": "A number from 0.7 to 1.2; 1.0 is normal speed."
            },
            "style": {
              "type": "number",
              "maximum": 1,
              "minimum": 0,
              "description": "A number from 0 to 1 controlling style exaggeration. A nonzero value adds processing time."
            },
            "stability": {
              "type": "number",
              "maximum": 1,
              "minimum": 0,
              "description": "A number from 0 to 1. Lower values allow more expressive variation."
            },
            "similarity_boost": {
              "type": "number",
              "maximum": 1,
              "minimum": 0,
              "description": "A number from 0 to 1 controlling how closely the generated speech follows the original voice."
            },
            "use_speaker_boost": {
              "type": "boolean",
              "description": "A boolean enabling extra similarity to the original speaker."
            }
          },
          "description": "An object containing any of the voice controls below. All controls can be omitted.",
          "additionalProperties": false
        }
      },
      "description": "The text to read aloud and optional voice settings.",
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}