Google launches Gemini 3.8 Flash TTS, debuts #2 in blind voice arena behind Sonic 3.6

ArtificialAnlys · x · 2026-09-23

Google DeepMind released Gemini 3.8 Flash TTS and Flash-Lite TTS, supporting preset voices and voice replication. In the blind-human Provider Voice Arena, Flash TTS debuts #2 with Elo 1,263 — just 9 points behind Sonic 3.6 (1,272) and ahead of Qwen-Audio-3.0-TTS-Plus (1,260). Flash-Lite debuts #6 at 1,236, both ahead of Gemini 3.1 Flash TTS (1,199). The models also top the Pronunciation Robustness Benchmark and Japanese Controlled Voice Arena.

Related event: Google DeepMind Launches Gemini 3.8 TTS Models with 30-Second Voice Cloning(16 posts)→

Original post →

More from Models

Models channel →