LMNT · Schema

LMNT Speech Synthesis Request

Schema for a text-to-speech synthesis request to the LMNT Speech API.

text-to-speechvoice synthesisvoice cloningaudio streamingconversational AIlow latencyreal-time audio

Properties

Name Type Description
voice string The ID of the voice to use for synthesis. Use the voices endpoint to list available voices.
text string The text to synthesize into speech.
format string Audio output format.
sample_rate integer Sample rate of the output audio in Hz.
speed number Speaking speed multiplier. 1.0 is normal speed.
temperature number Expressiveness of the synthesized speech. Higher values produce more expressive output.
quality string Audio quality level for synthesis.
language string BCP-47 language tag for the synthesis language. LMNT supports 31 languages.
return_timestamps boolean If true, the response includes word-level timestamps.
conversational boolean If true, optimizes for conversational speech patterns.
model string The LMNT model to use for synthesis. Defaults to the latest Blizzard model.
View JSON Schema on GitHub

JSON Schema

lmnt-speech-synthesis-schema.json Raw ↑ 
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://raw.githubusercontent.com/api-evangelist/lmnt/main/json-schema/lmnt-speech-synthesis-schema.json",
  "title": "LMNT Speech Synthesis Request",
  "description": "Schema for a text-to-speech synthesis request to the LMNT Speech API.",
  "type": "object",
  "required": ["voice", "text"],
  "properties": {
    "voice": {
      "type": "string",
      "description": "The ID of the voice to use for synthesis. Use the voices endpoint to list available voices."
    },
    "text": {
      "type": "string",
      "description": "The text to synthesize into speech.",
      "maxLength": 5000
    },
    "format": {
      "type": "string",
      "description": "Audio output format.",
      "enum": ["mp3", "wav", "webm", "aac"],
      "default": "mp3"
    },
    "sample_rate": {
      "type": "integer",
      "description": "Sample rate of the output audio in Hz.",
      "enum": [8000, 16000, 24000]
    },
    "speed": {
      "type": "number",
      "description": "Speaking speed multiplier. 1.0 is normal speed.",
      "minimum": 0.25,
      "maximum": 2.0,
      "default": 1.0
    },
    "temperature": {
      "type": "number",
      "description": "Expressiveness of the synthesized speech. Higher values produce more expressive output.",
      "minimum": 0.0,
      "maximum": 1.0
    },
    "quality": {
      "type": "string",
      "description": "Audio quality level for synthesis.",
      "enum": ["low", "medium", "high"]
    },
    "language": {
      "type": "string",
      "description": "BCP-47 language tag for the synthesis language. LMNT supports 31 languages.",
      "examples": ["en", "es", "fr", "de", "ja", "ko", "zh"]
    },
    "return_timestamps": {
      "type": "boolean",
      "description": "If true, the response includes word-level timestamps.",
      "default": false
    },
    "conversational": {
      "type": "boolean",
      "description": "If true, optimizes for conversational speech patterns.",
      "default": false
    },
    "model": {
      "type": "string",
      "description": "The LMNT model to use for synthesis. Defaults to the latest Blizzard model.",
      "examples": ["blizzard-2"]
    }
  }
}