OpenRouter STT icon

OpenRouter STT

Transcribe audio to text using OpenRouter speech-to-text models (Whisper and others). Accepts base64 audio or binary input.

OpenRouter STT

AI-generated

Summary

Transcribes audio from each incoming item using an OpenRouter speech-to-text model and returns the transcription with model and format details.

Inputs

  • Model (required) — OpenRouter STT model to use; choose a listed model or provide an ID.
  • Audio Source — Choose a binary property on the incoming item (default: data) or provide raw base64 audio. The base64 field is required for that source; a data URI prefix is stripped if present.
  • Audio Format (required) — Audio format: WAV, MP3, FLAC, M4A, OGG, WebM, AAC, or MP4 (default: MP3).
  • Language — Optional ISO-639-1 language code; omitted when blank, allowing automatic detection.
  • Temperature — Optional transcription sampling temperature from 0 to 1 (default: 0).

Output shape

One output item per input item, each containing a JSON object.

Successful items include text, model, and format; usage is included when returned by the API. Requires OpenRouter API credentials. If Continue On Fail is enabled, failed items contain an error field.

Links

Discussion