Create Speech | Mesh API Docs

curl -X POST https://api.meshapi.ai/v1/audio/speech \
     -H "Authorization: Bearer <token>" \
     -H "Content-Type: application/json" \
     -d '{
  "input": "Hello, welcome to our text-to-speech service. This is a sample conversion.",
  "model": "elevenlabs/eleven_turbo_v2_5",
  "voice": "Rachel",
  "stream": true,
  "response_format": "mp3",
  "language_code": "en-US",
  "voice_settings": {
    "stability": 0.75,
    "similarity_boost": 0.85,
    "style": 0.5,
    "use_speaker_boost": true,
    "speed": 1
  },
  "pronunciation_dictionary_locators": [
    {
      "pronunciation_dictionary_id": "dict12345",
      "version_id": "v1.0"
    }
  ],
  "seed": 42,
  "previous_text": "This is the previous sentence.",
  "next_text": "This is the next sentence.",
  "previous_request_ids": [
    "req-abc123",
    "req-def456"
  ],
  "next_request_ids": [
    "req-ghi789"
  ],
  "apply_text_normalization": "basic",
  "apply_language_text_normalization": true,
  "use_pvc_as_ivc": false,
  "enable_logging": true,
  "optimize_streaming_latency": 100,
  "speaker": "anushka",
  "target_language_code": "hi-IN",
  "pitch": 1.2,
  "pace": 0.9,
  "loudness": -3,
  "speech_sample_rate": 22050,
  "enable_preprocessing": true
}'

Convert text to speech.

Provider is resolved automatically from the model name via the model_pricing table — no hardcoded provider branching. Streaming is used when the provider adapter supports it and stream=true (the default).

The voice field is required for ElevenLabs models. Sarvam models use speaker instead.

Authentication

AuthorizationBearer

Bearer authentication of the form Bearer <token>, where token is your auth token.

Request

This endpoint expects an object.

inputstringRequired

modelstringOptionalDefaults to elevenlabs/eleven_turbo_v2_5

voicestring or nullOptional

streambooleanOptionalDefaults to true

response_formatstring or nullOptional

language_codestring or nullOptional

voice_settingsobject or nullOptional

pronunciation_dictionary_locatorslist of objects or nullOptional

seedinteger or nullOptional

previous_textstring or nullOptional

next_textstring or nullOptional

previous_request_idslist of strings or nullOptional

next_request_idslist of strings or nullOptional

apply_text_normalizationstring or nullOptional

apply_language_text_normalizationboolean or nullOptional

use_pvc_as_ivcboolean or nullOptional

enable_loggingboolean or nullOptional

optimize_streaming_latencyinteger or nullOptional

speakerstringOptionalDefaults to anushka

target_language_codestringOptionalDefaults to hi-IN

pitchdouble or nullOptional

pacedouble or nullOptional

loudnessdouble or nullOptional

speech_sample_rateinteger or nullOptional

enable_preprocessingboolean or nullOptional

Response

Successful Response

Errors

422

Unprocessable Entity Error