Qwen‑Audio‑3.0‑TTS offers real‑time and studio‑grade voice generation for Alibaba Cloud customers
The new text‑to‑speech model adds inline tag steering for emotions and breaths, and supports free‑form natural‑language prompts.
By Ryan Merket · Published
Primary source: X
Why it matters
Qwen‑Audio‑3.0 gives developers a versatile, low‑latency TTS option and signals Alibaba’s push to commercialise its AI stack against OpenAI and Google.
On July 23rd, 2026, Alibaba's Qwen research group announced the release of Qwen‑Audio‑3.0‑TTS on its official X account. The post – linked to the original tweet – introduced two deployment flavors: Flash, aimed at real‑time interaction, and Plus, which prioritises the highest generation quality.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.