Qwen-Audio-3.0 model leads global competition in three categories

Alibaba announced the launch of the Qwen-Audio-3.0 series on its artificial intelligence platforms. The audio model developed internally by the company achieved first place globally in three different categories in the latest evaluation by Artificial Analysis platform: speech recognition, speech synthesis, and real-time interaction. The series comprises three specialized solutions: one to convert audio to text with support for complex contexts and technical domains, another to generate speech from text in multiple languages with control over emotion and tone, and a third that enables continuous bidirectional dialogues with interruption capability and integration with other tools. Developers can access the functionalities through API calls or subscription plans.