The WAVE SUMMIT Deep Learning Developer Conference 2023, sponsored by the National Engineering Research Center for Deep Learning Technology and Application, and hosted by Baidu Flying Paddle and Wenxin Big Model, will meet you in Beijing on August 16. NVIDIA, as a technology partner, will join hands with Baidu Flying Paddle to participate in this technology event.
In this conference, NVIDIA and Flying Paddle jointly planned four major activities highlights, including NVIDIA Artificial Intelligence Workshop - NeMo Special Session, "Joining Flying Paddle Innovation Acceleration" theme sub-forum, "NVIDIA x Flying Paddle: Ultimate Performance, Empowering AI can be used in thousands of industries" keynote speech and a physical booth with wonderful demonstrations . At that time, the NVIDIA expert team will bring a wealth of keynote speeches and NVIDIA NeMo code practical learning opportunities, so that you can quickly master the latest large language model (LLM) technologies and solutions.
NVIDIA AI Workshop
August 16th 13:00-13:45
Hands-on experience with the voice AI development tool - NVIDIA NeMo code combat (offline participation only)
This work visit will focus on how to use NVIDIA NeMo to train a custom voice AI model, and at the same time experience the inference of the voice AI model in the edge computing device NVIDIA Jetson NX .
NVIDIA NeMo is an open source tool library for building conversational artificial intelligence applications. It uses speech and text data as input, and uses artificial intelligence and natural language processing models to understand semantics, so as to realize the mutual conversion between speech and text. NeMo toolkit can be used to build a conversational solution for voice and text interaction between humans and machines, and can be used to build application scenarios such as intelligent voice assistants, chat robots, intelligent voice translation, voice-controlled smart homes, and driverless car voice command interaction.
- Introducing NVIDIA NeMo
- Build a Speech Dataset for NeMo
- Using NeMo to train ASR speech recognition model method
- Using NeMo to train TTS speech synthesis model method
Hands-on experience: Implement voice AI model inference on Jetson NX edge computing device
NVIDIA theme sub-forum - join hands with flying paddles to accelerate innovation
August 16th 15:40 – 17:10
The development of large language models is an important trend leading the trend of the times. For LLM developers, excellent user experience, extreme performance, rich functions, stable platform and perfect customer service are all crucial concerns. As an important partner of Flying Paddle, NVIDIA has been based on the GPU hardware platform, helping Flying Paddle to advance together with Flying Paddle in multiple fields such as hardware adaptation, driver adaptation, acceleration library integration, and performance optimization.
round table dialogue
As the domestic LLM application has attracted much attention, more and more developers and engineers have begun to pay attention to the development technology of LLM. To this end, we invited Zhao Qiao, head of Baidu Flying Paddle & Wenxin Ecological Products, and Zheng Baicheng, Deputy General Manager of the Asia-Pacific Region of NVIDIA Development and Technology Department , to share with you what challenges you will face in the LLM development process, and how NVIDIA and Flying Paddle Work together to solve these problems. At the same time, the two will also talk about the development prospects of LLM technology and applications.
The Latest Practice of NGC Flying Paddle Container
Bin Zheng｜Senior Product Leader, NVIDIA Asia Pacific
While generative AI applications are flourishing, we have also noticed that more and more companies have chosen to develop their own LLM applications based on the domestic PaddlePaddle DLFW. This speech will share with you how NGC PaddlePaddle Container is adapted to the latest generation Hopper architecture GPU and the latest NVIDIA SDK software stack, and teach you how to easily obtain the best development environment, get rid of the limitation of environment adaptation, and ensure continuous upgrade of functions and performance. Help PaddlePaddle developers make full use of GPU performance and break through the traditional development framework.
Using NeMo Framework to Build E2E Large Model Solutions
Zhai Jian｜NVIDIA Senior Solution Architect
NVIDIA NeMo Framework is a large model solution launched by NVIDIA to build E2E. This speech will focus on the pain points that NeMo Framework can solve in large-scale model practice scenarios, the software stack structure adopted, and the various optimized parallel strategies adopted by the framework, so as to accelerate the entire E2E process. In addition, several successful cases of typical user scenarios will be shared. So that the audience can have a more comprehensive understanding of NVIDIA NeMo Framework.
NVIDIA AI Enterprise helps AI application development in the whole process
Xiaonan Cui｜NVIDIA Developer Ecosystem Development Manager
NVIDIA AI Enterprise is an end-to-end secure cloud-native AI software suite. As the operating system of NVIDIA AI, it can solve various challenges faced by customers from development to deployment while improving operational efficiency. With the help of a rich full-stack software library (Including AI solution workflow, framework, pre-training model and infrastructure optimization), combined with NVIDIA global technical support expert services, to help customers realize AI implementation and business value as quickly as possible.
"Intelligent Hardcore Ecological Co-creation" Parallel Forum Keynote Speech – NVIDIA×Flying Paddle: Extreme Performance, Empowering AI to Land in Thousands of Industries
August 16 16:40-16:50
Various applications of generative AI have emerged around the world. In the past six months, we have profoundly changed our way of life and greatly improved our production efficiency and work ability. By using the new NVIDIA hardware and software stack, it helps LLM developers reduce costs and increase efficiency, achieve technological innovation, and gain the trust of LLM enterprises through confidential computing functions.
Experience the NVIDIA physical booth and decipher the AI driving force behind the large model
NVIDIA empowers global enterprises to accelerate the construction of generative AI applications, launching a cloud service for creating large-scale language and vision models. Including the end-to-end secure cloud-native AI software suite from NVIDIA AI Enterprise , and how to customize text, image and biological cases through NVIDIA NeMo, Picasso and BioNeMo™ cloud services. At the booth, we will also show you NVIDIA NGC flying paddle container and model examples, so that you can understand how we provide developers with the best development environment.
On-site participants will also have the opportunity to get a small gift from NVIDIA, and partners who cannot be present can also watch wonderful speeches online through the online live broadcast of the conference, and participate in the full-stack acceleration journey of large-scale AI in real time!
