Tasks and inputs
Voxtral TTS is listed for Text-to-speech. It accepts text, audio and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
Voxtral TTS is Mistral's speech-generation model for narration and voice-conditioned output, released.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
Voxtral TTS is Mistral's speech-generation model for narration and voice-conditioned output, released.
The publisher documents nine languages, streaming output and zero-shot voice cloning without a transcript for the voice prompt. A voice prompt and the words to be spoken have different jobs: one conditions the delivery, while the other defines the content. Keep them separate in the application and in your evaluation notes.
Test numbers, names, abbreviations and words that change pronunciation across languages. A short greeting can sound convincing while a longer narration loses pacing or changes voice character. Listen through transitions and check the final recording in the same player and codec your audience will use.
For an interactive product, measure response onset on your own network and runtime. The publisher's latency figure is not a guarantee for a different host, input length or streaming pipeline.
The published weight licence is CC BY-NC 4.0. That restriction matters when planning a paid product. A managed API has its own service terms; do not assume the weight licence and the hosted contract are interchangeable. This page leaves ordinary text-token pricing unset because the speech output uses a different billing schedule.
Choose Gemini 3.8 Live when you need a bidirectional voice conversation, or use a TTS model when the transcript is already prepared.
Publisher source 1 Deployment limits and commercial terms belong to the selected checkpoint or API route.
Compare the model on the work you need to complete.
Voxtral TTS is listed for Text-to-speech. It accepts text, audio and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
Compare pronunciation, voice consistency and time to the first playable audio. Check the service’s billing unit for the actual script.
The weights are downloadable under CC BY-NC 4.0. Check runtime support and size the complete model and working context for your expected concurrency.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
Microsoft
MAI-Voice-2.1 generates expressive multilingual speech from text through Azure Speech, with voice selection and supported custom-voice workflows.
Microsoft
MAI-Voice-2.1-Flash generates expressive multilingual speech from text through Azure Speech, with voice selection and supported custom-voice workflows.
Gemini 3.8 Flash-Lite TTS converts supplied text into speech through the Gemini API. It is a candidate for read-aloud features and repeated single-speaker audio jobs.
Gemini 3.8 Flash TTS turns a supplied text transcript into audio. It is a speech-generation model for narration and dialogue, with speaker and style information supplied separately from the words to be spoken.
xAI
Grok Voice TTS 1.0 from xAI: Convert text into spoken audio with a single API call. The API supports a rich set of expressive voices, inline speech tags for fine-grained delivery control, and output formats from high-fidelity MP3 to telephony-optimized μ-law.
OpenAI
GPT-Realtime-2.1 from OpenAI: Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior.