Transcription on OpenRouter
Blog post from OpenRouter
OpenRouter provides a streamlined method for audio transcription by allowing users to send base64-encoded audio files to the POST /api/v1/audio/transcriptions endpoint, which returns a JSON response containing the transcribed text and usage details, such as duration and cost. This process utilizes the same API key and authentication as Chat Completions, eliminating the need for additional SDKs or services. The platform supports various speech-to-text models, including Whisper-class models that charge per audio second and newer models that charge per token. While users can send audio in formats like mp3, wav, and aac, the processing is limited to a 60-second upstream timeout, requiring longer audio to be split into segments. Although OpenRouter does not provide SRT/VTT output, users can obtain timestamps by setting the response_format to verbose_json, which works with OpenAI-compatible providers. The transcription service ensures cost transparency by reporting the exact cost per request in the usage object, with no markup added by OpenRouter.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.