Last refreshed 2026-10-16. Next refresh: weekly.
Why use Grok Voice Transcribe 2.0 on xAI Console?
xAI Console offers Grok Voice Transcribe 2.0. xAI, founded by Elon Musk in March 2023, focuses on advancing scientific discovery and understanding the universe through AI.
Setup recipe
Docs fallbackUse the provider REST API or SDKCreate a provider API keymodel: grok-voice-transcribe-2.0grok-voice-transcribe-2.0Request example
grok-voice-transcribe-2.0.Gotchas
No curated gotchas have been sourced for this exact provider/model route yet.
Capabilities
About Grok Voice Transcribe 2.0
Grok Voice Transcribe 2.0 (API id grok-voice-transcribe-2.0) is an xAI speech-to-text / ASR model, announced September 18, 2026 on x.ai/news. Specialized transcription API, not a chat LLM. First-party: twice as accurate as Grok Voice Transcribe 1.0 at the same price; multilingual (dozens of languages, auto language detect, mid-recording language switch); batch and streaming STT; word-level timestamps with confidence; speaker diarization at no extra cost; multichannel up to 8 channels; key-term biasing (up to 100 terms); text formatting for numbers/dates/currencies/phones/emails. Default model when model is omitted on the Speech-to-Text API; pin grok-voice-transcribe-1.0 to stay on the prior version during deprecation.