Recently, the project "Research and Industrialization of Motor Intelligent Detection Platform Based on AI Voiceprint Feature Analysis" independently developed by Osen was officially awarded the "Certificate of Scientific and Technological Achievements" (No. 3050 of 2025 of the China Private Science and Technology Promotion Association) after review by the China Private Science and Technology Promotion Association. Its AI voiceprint intelligent detection technology has been evaluated as "internationally advanced level", filling the domestic gap and marking that the company's technical strength in the field of AI and acoustic detection has once again been recognized by national authorities.

The evaluation committee for this event is composed of five experts and scholars, including Professor Ding Tianhuai from Tsinghua University, Professor Wu Mingxin from Beihang University, the General Research Institute for Machinery Science, the Shenzhen Institutes of Advanced Technology of the Chinese Academy of Sciences, and the Torch Center of the Ministry of Science and Technology.

The evaluation committee listened to the technical report, reviewed the evaluation materials, and after questioning and discussion, unanimously agreed that this project fills the domestic gap in the field of small and micro motor abnormal noise detection technology and equipment, and its AI voiceprint feature analysis motor abnormal noise intelligent detection technology has reached the international advanced level.

The high-end and intelligent development demands of modern industrial technology have injected new impetus into the field of motor drive, but also posed more severe challenges to the operation and control of motor systems. Motor fault detection, as a key technology to ensure the safe operation of motor equipment, plays an irreplaceable and significant role in the reliable control and intelligent operation and maintenance of motor systems.

At present, although traditional fault diagnosis methods can identify motor faults to a certain extent, they have limitations such as low detection efficiency and significant influence of human factors on accuracy. In response to the detection requirements of small and micro motors, Osens has independently developed an intelligent motor detection platform based on AI voiceprint feature analysis. Compared with traditional algorithms, the yield rate has been increased to 98.5%, and real-time detection functions have been realized. The detection time has been shortened to 10ms, achieving intelligent and efficient motor detection.

Multi-dimensional integrated collection

Innovate and develop the motor operation sound spectrum acquisition technology. By integrating digital signals of rotational speed, current, vibration and yaw, a characteristic sound spectrum set of motor operation is formed.

Significantly improve the detection accuracy

The sound feature extraction and recognition technology based on physical neural network models and multi-layer convolutional neural networks significantly improves the accuracy of motor abnormal noise detection compared with traditional detection methods.

Meet the testing requirements of all scenarios

It meets the full-scenario testing requirements of motors under no-load, load, radial and axial forces, and has excellent scalability. It can be applied to motor testing in different fields such as white goods, new energy vehicles and smart devices.

In the future, Osens will always adhere to innovation-driven development, further increase investment in research and development, explore application potential in more fields, continuously deepen the transformation and application of innovation achievements, empower the high-end equipment manufacturing industry to move towards intelligence and digitalization, and contribute more to the high-quality development of "Made in China".