Alibaba Cloud launches Qwen-Audio-3.0, tops global voice AI rankings

Alibaba Cloud has launched the Qwen-Audio-3.0 series of voice models, including speech recognition (ASR), text-to-speech (TTS), and real-time interaction (Realtime). They're now available on the Qwen platform and Alibaba Cloud Bailian, with API access and Token Plan options. The series topped all three categories in Artificial Analysis's July voice rankings. ASR handles complex and professional speech, TTS supports multiple languages and dialects with emotional control, and Realtime enables end-to-end voice conversations with interruptions and tool calls. The models are already used in Qwen App, Qwen Office, and Qoder.