xAI TTS v1

• 🗣️ xAI's text-to-speech model, converting written text into spoken audio.
• 🆕 First TTS entry in xAI's audio model lineup.
• 🔧 Accessed via a dedicated text-to-speech API endpoint.
• 🎯 Takes text plus a voice identifier and language parameter.
• 💬 Offers multiple named voices, including one called "eve".
• 🔒 Requires an xAI API key for direct access.

xAI

xAI is an American artificial intelligence company and wholly owned subsidiary of SpaceX. The company develops AI systems under the Grok brand, spanning language models, image generation, and video synthesis. xAI has quickly established itself as a multimodal…

xAI TTS v1 is the text-to-speech model from xAI, the AI company behind the Grok family of products. Released in 2026, it converts written text into natural-sounding spoken audio through xAI's audio API, accepting a text string, a voice identifier such as "eve", and a language parameter, and returning generated audio.

As xAI's first dedicated TTS model, there is no same-family predecessor to compare against; it instead expands xAI's modality coverage alongside its companion xAI Speech to Text v1 released around the same time, together forming a speech input/output pair that complements the text models in the Grok 4.3 line. These audio capabilities make it possible to build voice-driven pipelines around Grok.

Developers can call the model directly via xAI's API endpoint, and a voices listing endpoint is provided to enumerate the available voice identifiers and names. Direct access requires an xAI API key.

Because xAI has not published a detailed technical report or benchmark results for this model in the sources available here, specifics such as supported language count, sampling rate, or architecture are not documented publicly. This description is therefore limited to the verified API behavior detailed in xAI's own documentation.