OpenAI-compatible Audio API

Configure the OpenAI SDK with your Kitsch API key, Audio API URL, and model to call the audio APIs.

SDK connection

Set KITSCH_OPENAI_BASE_URL to your Audio API URL ending in /v1, and KITSCH_API_KEY to an active workspace API key.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["KITSCH_API_KEY"],
    base_url=os.environ["KITSCH_OPENAI_BASE_URL"],
    max_retries=0,
)

API documentation

• Speech API
• Speech Streaming API
• Transcriptions API
• Transcription Streaming API
• Realtime Transcription API

Errors and retries

Errors contain error.message, error.type, error.param, and error.code. Include x-request-id when contacting support. Disable automatic retries: resubmitting after a timeout or disconnect may result in duplicate usage.