Gemini 3.1 TTS Preview (demo on Google Ai studio https://aistudio.google.com/generate-speech?model=gemini-3.1-flash-tts-preview) is Google’s latest text-to-voice model.

Two modes of presentation: text or “compose” which allows the separation of speakers.

Multi-speaker support, 32 voices available with styles, bitrates, and accents for English-language texts.

Fine management of expressions with “emotions” that can be added throughout the text, on a word as well as a block of text.

Generations up to 1 minute 20.
Export mp3.

Free and seems unlimited.


Source: Art poétique, Verlaine

Source: Les fourberies de Scapin, Molière