Gemini 3.8 text to speech
September 27th, 2026
This actually sounds kinda… good? I generated it with Google’s new Gemini 3.8 Flash text to speech model. It’s narrating the first paragraph from Kenneth Grahame’s “Wind in the Willows” (one of my childhood favourites). I have a never ending list of things to read, and a morning and afternoon commute where the last thing I feel like doing is looking at yet another screen. I think I could actually listen to this.
Voice: Algenib
PromptRead this as a polished nonfiction audiobook narrated by a warm, intelligent British speaker. Use a natural contemporary neutral southern English accent - clear, elegant, and authentic, never theatrical, exaggerated, or caricatured.
Speak with calm confidence and a conversational tone, as though explaining the subject to an interested listener. Use measured pacing, clear articulation, and natural variation in pitch and emphasis. Slow slightly for important ideas, definitions, figures, and transitions. Add brief pauses between sections and after major points.
Pronounce names and place names carefully and consistently. Use British pronunciation and vocabulary where appropriate, but do not rewrite the author’s wording unnecessarily.
The overall result should sound like a professionally produced British nonfiction audiobook: engaging, thoughtful, accessible, and easy to follow over a long listening session.