Google ships two speech-synthesis models; top rankings are vendor-cited
Gemini 3.8 Flash TTS and Flash-Lite TTS are now live on Google Cloud; the quality-leading claim is vendor-cited, so treat it as a product signal.
ImportanceLocalEvidenceE2 unreplicatedWrite-upQuick
Google made two new text-to-speech models, Gemini 3.8 Flash TTS and Flash-Lite TTS, available on its cloud platform on September 23. Flash TTS targets audio quality; Flash-Lite TTS targets cost and inference speed.
Flash TTS supports 130 languages at launch, Flash-Lite TTS 101. Both share similar APIs, offer more than 2,000 prepackaged voices, and can clone a voice from a 30-second sample, with Google requiring developer-secured speaker consent. Generated audio carries a SynthID watermark and a C2PA record.
The benchmark-topping claim comes from the Hume AI audio-quality benchmark and Voice Arena human-preference evaluations as cited by Google; it is a vendor-cited result, not independently verified.