OpenRouter Speech-to-Text Transcribe audio via using . Requires (get one at https://openrouter.ai/keys). If unset, stop and ask. This endpoint is not OpenAI-compatible. The body is JSON with base64 audio under — not with a field the way OpenAI's works. Do not point the OpenAI SDK at this endpoint; it will send the wrong shape. Use , , or directly. One call, JSON back Both request and response are JSON. The response body carries: - — the transcript. - — always includes . Providers additionally report either of audio billed or a token breakdown ( , , ), depending on how they price the request.…