xAI · 2026-07-06 · major
xAI Grok Voice — 21 new flagship voices with speech tags for pacing
xAI added 21 new multilingual voices to Grok Voice, upgraded the original five, and shipped inline speech tags like [pause] and <whisper>. The voices work across 25+ languages via the Voice Agent API, TTS API, and Voice Agent Builder.

Grok Voice now ships 21 new multilingual flagship voices plus inline speech tags for pacing and whispering.
Quick facts
| Maker | xAI |
|---|---|
| Voices | 21 new + 5 upgraded |
| Languages | 25+ with mid-conversation switching |
| Availability | Voice Agent API, TTS API, Voice Agent Builder |
| What's new | Inline speech tags like [pause] and <whisper> |
| Voice roles | Support, characters, commentary, advertising, education |
What is it?
Grok Voice's catalog jumped to 21 new flagship voices — including Lumen, Castor, Naksh, Atlas, Carina, Zagan, Helix, Orion, and Luna — plus dedicated support, wellness, and character voices, on top of upgraded versions of the original five. Every voice supports 25+ languages with mid-conversation language switching.
How does it work?
Each Grok voice is a distinct TTS profile tuned for pacing, phrasing, and prosody, so builders cast a voice per job (customer support, characters, commentary, advertising, education) rather than reusing one narrator. New inline speech tags like [pause] and <whisper> act as directives the voice model renders at synthesis time, so a single script can control timing and emphasis without swapping voices.
Why does it matter?
The Grok Voice roster is finally large enough to give each product persona its own voice instead of forcing every use case into the same narrator. Speech tags close the gap with professional voiceover — the same voice can whisper a warning, hold a beat before a punchline, or switch language mid-sentence — which is the kind of control voice-agent products have been asking for.
Who is it for?
Voice-agent builders, telephony and support teams, character and ad-tech media producers.
Frequently asked questions
- How many new Grok voices did xAI add?
- xAI added 21 new flagship voices to Grok Voice on 2026-07-06 and upgraded the original five for pacing and phrasing. Named additions include Lumen, Castor, Naksh, Atlas, Carina, Zagan, Helix, Orion, and Luna, along with dedicated support, wellness, and character voices developers cast per job.
- What speech tags does Grok Voice support?
- Grok Voice now reads inline speech tags at synthesis time. Directives like [pause] and <whisper> let a script control timing and emphasis without switching voices, and every voice can shift language mid-conversation across the 25+ supported languages.
- Where can developers use the new Grok voices?
- The new Grok voices are available through the xAI Voice Agent API, the TTS API, and the no-code Voice Agent Builder xAI released earlier this week. All three surfaces expose the same voice catalog and the new speech tags.
- How is this different from the Voice Agent Builder launch?
- The Voice Agent Builder is the no-code harness for wiring a voice agent's tools, prompts, and telephony. This release expands what those agents can sound like: 21 new voices, five upgraded ones, and new speech tags, all cast through the same builder and APIs.
Try it
Cast a voice via the Voice Agent Builder or pass voice= in the Voice Agent API — see docs.x.ai/docs/guides/voice.