[Alibaba Turns AI Voice into Productivity Suite: Recording, Agent, Podcast All-in-One] According to monitoring by Beating, Alibaba has launched a one-stop AI voice platform called CosyVoice Studio, powered by the Qwen-Audio series models. It integrates voice recording, real-time voice agents, and audio creation in one go. CosyFlow focuses on voice recording. When users speak, it automatically removes filler words like 'um,' 'ah,' and 'that,' as well as repetitive expressions. It can also organize fragmented speech directly into lists, emails, and meeting minutes. During recording, it can distinguish between different speakers, automatically generate summaries, and create to-do lists. CosyAgent is designed for enterprise voice services. Users can create real-time conversational agents using natural language and integrate them with enterprise documents, databases, MCP, and APIs for scenarios like customer service and telemarketing. CosyCreative handles audio production. By uploading documents, webpages, or inputting text, users can generate podcasts and multi-character audiobooks. It also supports voice cloning, partial regeneration, and multi-track editing. The personal version is now available on iOS, Android, Mac, and Windows, with a limited-time free trial. CosyAgent and CosyCreative are currently in whitelist testing for enterprises. [Original Link]