Building blocks

Text-to-speech

Turn a script into speech audio with one of your cloned voices. Useful on its own, and the same engine video jobs use internally when you pass them a script.

Synthesize speech

Costs 10 credits per 1,000 characters (minimum 10), charged immediately. The voice_id must be one of your voices — someone else's id returns 404. The speech engine is fixed; there's nothing to configure.

curl -X POST https://platform.promoat.ai/api/partner/v1/tts \
  -H "Authorization: Bearer $PROMOAT_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "voice_id": "VOICE_ID",
    "text": "Welcome back! Here is what is new this week.",
    "language_code": "en",
    "voice_settings": {
      "stability": 0.4,
      "similarity_boost": 0.75,
      "style": 0.3,
      "speed": 1.1,
      "use_speaker_boost": true
    }
  }'

{
  "audio_url": "https://…/tts-audio.mp3",
  "duration_sec": 4.02,
  "asset_id": "…",
  "credits_charged": 10
}

When to call this directly

• Preview a voice before committing a video job to it — 10 credits to hear the clone say your actual script.
• Audio-only products — voice notes, IVR lines, podcast inserts.
• Pre-generating narration you'll reuse: pass the resulting audio_url to a talking-head video instead of a script, so multiple videos share the exact same take.

For a normal one-off video you can skip this endpoint entirely — POST /videos with a script runs TTS inside the job.

Try it

Grab a voice_id from the list first, then paste it into the body: