Gemini TTS (https://aistudio.google.com/generate-speech) is Gemini’s text-to-voice model, developed in two versions, Gemini-2.5-pro-preview-tts and gemini-2.5-flash-preview-tts, and available on Google ai studio.

One or two voices in the form of dialogue can be generated and the tones specified by nuances that are accounted.
During the tests, French very well taken care of.

30 voices available (of high quality despite some a few a little “metallic”), export in WAV format.

“Preview” version, free and seems unlimited.

Sources for testing:

• Art poétique Verlaine, https://fr.wikipedia.org/wiki/Art_po%C3%A9tique_ (Verlaine)
• The Miser, Molière, Act III scene 4 and Act IV scene 5, https://libretheatre.fr/lavare-de-moliere/