Agent Integrations

Sarvam speech-to-text (Saaras) for 22 Indian languages and English

Transcribes (or, with mode=translate, translates to English) up to 30 seconds of audio. Billed per second of audio at Sarvam's list price (₹30/hour, converted at SARVAM_INR_PER_USD) plus AGENT_INTEGRATION_SARVAM_MARGIN. Duration is measured from the file (PCM WAV from its byte rate), floored by its size and by the end of the last word Sarvam transcribes. The response is Sarvam's plus durationSeconds and costUsd.

post/agent-integrations/sarvam/speech-to-text

Request

  • Base URL: https://api.tinyhumans.ai
  • URL: https://api.tinyhumans.ai/agent-integrations/sarvam/speech-to-text
  • Auth: HTTP bearer

Response

Transcript (transcript, language_code, timestamps?, request_id) plus durationSeconds and costUsd

Changes

No recorded changes to this operation across all 1 revision of this API.