Alibaba today announced the launch of China’s first all-in-one AI voice productivity platform, CosyVoice Studio.
According to the company, the platform is built on its self-developed Qwen-Audio speech model and includes CosyFlow for voice recording, CosyAgent voice agents, and CosyCreative audio content creation tools.
Users can search for “CosyVoice” in the iOS, Android, Mac, and Windows app stores to download the app and try it free for a limited time. CosyAgent and CosyCreative are currently available to enterprise users for testing by invitation only and will be fully open soon.
CosyVoice Studio features:
Voice recording builds semantic understanding and text generation capabilities on top of speech-to-text conversion. In everyday scenarios, it automatically filters out verbal fillers such as “um” and “uh” and organizes scattered ideas into clear expressions. In work scenarios, users can dictate emails, work reports, and meeting minutes with appropriate formats and wording generated automatically.
The built-in “Quick Notes” feature is designed for meetings, interviews, classes, and other scenarios, supporting real-time recording and offline uploads. It distinguishes speakers by their voiceprints and automatically generates structured chapters and intelligent summaries after recording ends.
CosyAgent supports quickly creating agents with enterprise knowledge, tool-calling, and real-time voice interaction capabilities through natural language. It is also compatible with prompt and workflow configurations for scenarios such as intelligent customer service and telemarketing.
CosyCreative includes thousands of voices and supports voice cloning, producing voices that sound highly similar to real people. Users can upload documents, provide webpage links, or enter text to generate conversational podcasts or multi-character audiobooks, covering the full audio content creation workflow.
