Inworld · Speech
TTS 2 Flash
Latency-first realtime voice synthesis holding one voice identity across 200+ languages, tuned for high-volume streaming workloads.
Evidence and sources
Released 9 Aug 2026Maker HQ: US (source)
Sources: EmpirioLabs data, llminfo researchPrimary data from EmpirioLabs
from llminfo research: release date
Input & output
from the model's typeText
Audio
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works