DeepSeek has released a new set of API capabilities that grant developers access to the same toolset powering its consumer application, including voice interaction, web search, and file parsing. To facilitate the transition, the company provided an open-source frontend demo and a method for importing conversation history, allowing third parties to build their own versions of the DeepSeek app experience.
The updated API provides standalone access to optical character recognition (OCR), automatic speech recognition (ASR), and text-to-speech (TTS) capabilities. In a statement, DeepSeek said it wants developers to use the infrastructure to avoid repeated basic work and create more diverse products, noting that its own existing conversation product is a relatively basic form of the technology.
