Audio

Voices

GET /v1/audio/voices

Returns available text-to-speech voices.

Speech to text

POST /v1/audio/transcriptions

curl https://api.parel.cloud/v1/audio/transcriptions \
  -H "Authorization: Bearer pk-dev-YOUR_KEY" \
  -F model="whisper-large-v3-turbo" \
  -F file="@audio.mp3"

Text to speech

POST /v1/audio/speech

speech = client.audio.speech.create(
    model="elevenlabs-multilingual",
    input="Merhaba!",
    voice="alloy",
)

The response is an audio stream.

Music generation

POST /v1/audio/music

{
  "model": "elevenlabs-music",
  "prompt": "short upbeat synthwave intro",
  "duration": 15
}