Focusing on new AI technologies, HMS Core machine learning service injects new impetus into the intelligence of mobile applications

In recent years, artificial intelligence technology (hereinafter referred to as AI technology) represented by machine learning has developed vigorously. New algorithms emerge in an endless stream, and the developed capabilities such as image recognition, natural language, and live detection make an intelligent future life no longer out of reach. At the same time, these AI technologies are continuing to evolve and develop, and the limitations of data and computing power are constantly being broken through. Relying on the emerging new technologies and products, various applications, such as transportation, shopping, express delivery, and financial management, which are closely related to users' lives, have been implemented, becoming the focus and force of developers.

Simply put, machine learning is a branch of AI technology, while deep learning is an important branch of machine learning. HMS Core Machine Learning Service (ML Kit) provides developers with easy-to-use, diverse services, and leading-edge machine learning capabilities, helping developers to develop various AI applications faster and better.

Machine learning multi-domain, all-round empowering developers, welcome new breakthroughs in applications

Financial Transactions and Security

Thanks to the convenience of Internet finance, users can use mobile devices to complete transfers, wealth management and other services. However, the security of user data and assets has always been questioned. By training machine learning models to detect activities that may be fraudulent, a risk-controlled transaction process can be achieved.

To make the service faster and more secure, the HMS Core machine learning service can work with various user authentication systems, such as biometrics, to identify and authenticate the process. The capabilities of face detection, liveness detection, and card recognition in HMS Core's machine learning service are a good option for mobile apps that require strict user consent before any transaction is processed.

Intelligent Human-Computer Interaction

The combination of speech recognition and speech synthesis enables people to use their hands free, with the help of machine learning, to open up new app experiences through the interaction of voice commands. Applications such as smart home and smart customer service have now provided us with a variety of services in our lives.

Once a developer's app integrates machine learning services, it can become smart enough. When the user issues a voice command, the real-time speech recognition capability of the HMS Core machine learning service can realize the real-time recognition of audio streams in milliseconds, and the recognition accuracy rate measured in the laboratory is as high as 95%. The speech synthesis capability can directly convert text into natural and smooth human voice, allowing applications and devices to "speak", and the user experience is more personalized.

cross-language communication

From overseas travel, to international cultural exchanges and foreign trade, language barriers are a natural pain point. With the increase of domestic enterprises and application overseas business, this problem needs to be solved urgently. HMS Core machine learning service can realize text translation in source language and target language, and can translate foreign language signs, menus, manuals, etc. in real time through mobile phone photography.

In the face of conference scenarios with diverse fields, complex requirements and high real-time performance, traditional text translation cannot meet the high real-time requirements of simultaneous interpretation. Based on this, HMS Core machine learning service will soon launch a comprehensive and efficient machine simultaneous interpretation solution. . Simultaneous interpretation integrates the capabilities of real-time speech recognition, translation, and speech synthesis in the HMS Core machine learning service, supports real-time translation of audio streams into texts in different languages, and outputs audio content in multiple languages, thus balancing delay and accuracy. , to achieve effective multilingual communication at low cost.

digital image processing

Images are an important way for people to describe and express the representations and properties of things. Scenarios such as webcasting, online education, and video conferencing are booming, and image processing is becoming more and more important. The development of scenarios also puts forward higher requirements for technology. As a result, machine learning is increasingly applied to real-time audio and video scenarios, such as super-resolution, beauty, and image segmentation. The HMS Core machine learning service can blur the background of the video or replace the background of the video through the portrait segmentation technology, which can greatly reduce interference while protecting privacy.

As a magical technology to make blurred images clear, image super-resolution is one of the very important technologies in the field of image enhancement. Whether it is an early classic movie or an old photo, the deep learning-based super-resolution algorithm can complete the high-definition reconstruction of multimedia content without consuming a lot of resources. The whole process takes less time, but the actual effect is greatly improved. In addition, in the fields of medical imaging, transportation systems, and autonomous driving, image processing technology is also applied to varying degrees.

Machine learning + N, providing multi-dimensional industry solutions

In addition to the above-mentioned application scenarios, HMS Core machine learning services enable capabilities in many fields, continue to open up AI technology to help developers innovate applications, benefit users' work and life, meet more personalized customization demands, and further enhance application intelligence.

Machine learning + situational awareness to enhance the refined application experience

AI does not tell users to learn how to operate, but adapts to user needs to provide accurate and efficient services. How to perceive user needs faster and more accurately? Know when and where they need which service? This is one of many challenges that developers and operators face as they continue to improve their applications.

Relying on machine learning services, HMS Core Awareness Kit (Awareness Kit) supports more refined experience perception, enriching situational awareness from five dimensions: user status recognition, user behavior recognition, device status recognition, human-computer interaction perception, and surrounding environment perception. capabilities, and call these capabilities to help apps understand the user's current situation more quickly and efficiently, so as to provide a more intelligent and considerate experience.

Travel apps can remind users of the weather conditions at the destination, prepare for travel, and provide intimate travel companion services; music apps can link headphones or car Bluetooth to provide personalized music recommendations based on user activity status or time. , such as night driving music, etc.; and life service apps can accurately push relevant service information for users when they approach the target area and stay for a certain period of time.

Machine learning + HiAI Foundation, HiAI Engine, make the application more intelligent and powerful

Faced with changing application scenarios, the HMS Core machine learning service continues to optimize model lightweighting, model accuracy, and model occupancy, and is committed to providing developers with comprehensive and in-depth Huawei machine learning open capabilities. After the application integrates the HMS Core machine learning service, MindSpore Lite provides a full-scenario AI inference framework with easy-to-use interfaces, complete operator performance, and extensive support for third-party models.

HUAWEI HiAI Foundation, which opens the door to a new world for developers, has launched features such as device-cloud collaboration, Model Zoo, model quantification toolkit, network structure search toolkit, and multi-IP in-depth collaboration, continuously empowering mobile phones, tablets, smart screens, and car technology. , watches and other terminal devices, creating a computing environment with higher performance and lower power consumption, helping developers to quickly develop and deploy AI applications.

HUAWEI HiAI Engine has open application capabilities and supports the integration of various AI capabilities and applications, including CV engine, which can simulate the human visual system to perceive the surrounding environment, judge, identify, and understand the composition of space; ASR engine-transform human voice into The text is convenient for the computer to further parse and understand. NLU Engine - Allows computers to understand human voice or text for communication or natural movements.

The rapid development of computers and networks has made machine learning more and more important, and it has also changed people's lives and work. In the future, facing the refined demands of users and the intelligent upgrade of mobile applications, developers will need more and better machine learning technologies to realize application innovation and bring users a more intelligent experience. The HMS Core machine learning service will also move forward with this goal. Welcome to the official website of Huawei Developer Alliance to learn more about the technical capabilities of HMS Core AI, as well as more innovative and open capabilities.

Learn more details>>

Visit the official website of Huawei Developer Alliance
Get development guidance documents
Huawei Mobile Services Open Source Warehouse Address: GitHub , Gitee

Follow us to know the latest technical information of HMS Core for the first time~