Products / Grok Voice

grok-voice

Grok Voice

The Voice API covers speech-to-speech, text-to-speech, and speech-to-text. Voice Think Fast 2.0 was introduced Jul 29, 2026. Speech-to-speech starts at $0.05 / min; TTS is $15.00 / 1M characters. STT gained an adjustable VAD threshold on Jul 23, 2026.
id
grok-voice
current_version
think-fast-2.0
last_updated
generated_at
published_at (this UI deploy)
SpaceXAIGAX Stack
ProductDocs

Features

  • Speech to speech

    Realtime voice agents. Starting at $0.05 / min.

    voiceapirealtime

  • Text to speech

    TTS at $15.00 / 1M characters.

    voiceapi

  • Speech to text

    STT with adjustable VAD threshold (vad_threshold) for streaming and batch.

    voiceapi

  • Voice Think Fast 2.0

    Introduced Jul 29, 2026. grok-voice-latest routes to grok-voice-think-fast-2.0 starting August 5, 2026.

    voice

  • Custom voices

    Clone a voice from a reference clip (max 120s) via /v1/custom-voices and use the returned voice_id for TTS and speech-to-speech.

    voiceapi

Changelog

  • Release

    Introducing Grok Voice Think Fast 2.0

    can · Voice Think Fast is 2.0

    Grok Voice Think Fast 2.0 announced as the next-generation speech-to-speech voice model.

  • Feature

    Adjustable VAD threshold

    can · STT accepts vad_threshold

    Speech-to-text accepts vad_threshold as a streaming query param and batch multipart field. 0 disables the gate.

  • Feature

    21 New Flagship Grok Voices

    can · Grok Voice has 21 new flagship voices

    21 new multilingual flagship voices for Grok Voice, plus improved naturalness for the original five, in the Voice Agent API, TTS API, and Voice Agent Builder.

  • Feature

    Introducing the Voice Agent Builder

    can · Voice Agent Builder is in beta

    Voice Agent Builder beta: no-code production voice agents on Grok Voice, with telephony, tools, MCP, and observability.