Audio + speech with Meridian

Convert text into lifelike spoken audio. Choose from hundreds of neural voices across dozens of languages and dialects.

Request body

{
  "text": "Hello from Meridian.",
  "voice": "en-US-AvaMultilingualNeural",
  "format": "mp3"
}

Response

Returns audio/mpeg binary stream. Set the Accept header to match your requested format.

Transcribe audio files into text with optional diarization, punctuation, and profanity filtering.

Request

Content-Type: multipart/form-data

file:        @recording.wav
language:    en-US
diarize:     true

Response

{
  "text": "Full transcript text here.",
  "segments": [
    {
      "speaker": "Speaker 1",
      "text": "Hello from Meridian.",
      "start": 0.0,
      "end": 2.3
    }
  ]
}