Grok STT is xAI's speech-to-text: send an audio URL and get back the transcript as text plus word-level timings and the detected language. It covers 25 languages, with per-channel transcripts and key-term hints.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
- No sample yet
split-channel-call A 12-second stereo recording with a woman on the left channel and a man on the right, transcribed with multichannel on. Each side comes back on its own line; with multichannel off, the same file came back with only the woman's words.
- multichannel
- No sample yet
names-with-keyterms A lounge announcement with two invented names, sent with both names as key terms. Without them the same audio came back as "Xandrel" and "Corvessa".
- No sample yet
booking-numbers A spoken booking confirmation with language set to en and text_normalization on: the price and date come back as "$1,250" and "March 14" instead of words.
- en
- normalization
This example starts at $0.00003 per output second. The Playground updates the estimate when you change the settings.