DeepPavlov is an open-source framework for building chatbots and virtual assistants. It comes with a set of predefined components for solving Natural Language Processing (NLP) related problems and a framework for building a modular pipeline. This lets developers and NLP researchers create production-ready conversational skills and complex multi-skill conversational assistants.
DeepPavlov is used across many areas, including process automation of call centers and customer service, question answering systems at scale, sentiment analysis of customer reviews, production-ready dialog systems, and applied NLP research.
The DeepPavlov container consists of pre-trained models that use BERT and other state-of-the-art deep learning models for classification, NET, Question-Answering, and NLP tasks. DP Agent, a powerful orchestrator that reuses the declarative approach of the DP Library to form pipelines and build conversational AI experiences in a modular system. You can run your own pre-trained models using Python code, command line interface, APIs, or Docker.
With up to 20X speedups when running ASR/TTS pipelines on a V100 GPU vs CPU, DeepPavlov can help accelerate your NLP applications.
“DeepPavlov conversational AI technology is packed in an easy-to-deploy GPU-optimized container hosted on NGC to empower developers around the world to build production-ready, scalable, and reliable solutions as fast as never before,” said Mikhail Burtsev, Founder and Leader, DeepPavlov.ai. “Becoming a part of the NGC family was a short and smooth process for us.”
Get started by downloading the GPU-optimized DeepPavlov container from NGC.
Learn how to build a simple AI assistant with DeepPavlov and NVIDIA NeMo, on the NVIDIA Developer Blog.
