Voice & speech
ElevenLabs, OpenAI TTS, VoxCPM2 and dots.tts from the same key.
Steps
01
List voices
GET /v1/audio/voices returns engines and their voices.
02
Generate
POST /v1/audio/speech with model, voice and input text.
03
Save the audio
The response is the audio file itself.
bash
curl https://api.luxechain.si/v1/audio/speech \
-H "Authorization: Bearer lxc_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai-tts",
"voice": "alloy",
"input": "Welcome to LuxeChain.",
"response_format": "mp3"
}' --output speech.mp3
Speech is charged per 1,000 characters, roughly one minute of audio. Voice cloning is available on engines that support it.