:::

【2020 Solutions】 RelaJet Envisions the Future to Aid Hearing-Impaired with Easy 'Listen & Speak'

For the hearing-impaired, being able to enjoy a low-noise, good listening experience at a relatively affordable price is perhaps the greatest happiness in life. Chen Bai-ru, the founder of Future Insight Tech and a hearing-impaired individual, 'heard' the voices of the hearing-impaired. Understanding the struggles of not being able to hear clearly, he utilized AI technology to develop a 'Multi-person Voice Separation Engine' to help solve their challenges.

The Opportunity to Hear Again

According to statistics, there are about 99,535 people with hearing disabilities in Taiwan, accounting for about 10% of those with disabilities. These nearly 100,000 hearing-impaired individuals live in a world where 'listening and speaking freely' is challenging.

For the hearing-impaired, the two main problems faced are:

One, hearing aids are expensive; among the six major global brands, the average price is around NT$60,000, with high-end models even reaching NT$150,000, which is not feasible for an average middle-class family;

Two, the performance of traditional hearing aids is not sufficient; when the surrounding environment is too noisy or the volume too high, it becomes very difficult to clearly hear the speaker's voice.

Due to their size, conventional noise cancellation methods involving several microphones cannot be used in typical Bluetooth headsets. In noisy environments with many people talking, such as restaurants, gyms, and supermarkets, the quality of noise reduction during calls is not ideal.

Therefore, Chen Bai-ru, also a hearing-impaired individual and the founder of Future Insight Tech, uses AI deep learning technology to achieve noise removal and output clean human voices with just a single microphone.

Feature Recognition Completed in 10 Milliseconds with No Latency in Speaking and Hearing

Being able to complete all feature recognition calculations within 10 milliseconds is the greatest advantage of the RelaJet Multi-person Voice Separation Engine. Why 10 milliseconds? Because if the processing time for speech by hearing aids exceeds this limit, it can lead to a delay that causes dizziness in the individual. Therefore, any hearing aids classified as medical devices are required to complete all processing steps within 10 milliseconds.

RelaJet Core Technology - Voice Separation

The US Food and Drug Administration (FDA) is set to allow the sale of non-prescription (Over-the-Counter, OTC) hearing aids in 2020, which will significantly reduce the costs of experimentation and certification and make hearing aids more affordable. Additionally, the purchasing channels will be more open, eliminating the cumbersome fitting process. Future Insight is seizing this business opportunity, actively establishing partnerships with the top six global hearing aid brands, and also entering the Bluetooth headset market to benefit more hearing-impaired individuals.

Future Insight Tech's specific approach to incorporating AI involves a customized model with an algorithm chip, achieving a noise reduction of 20Db and power consumption below 9Ma. This noise reduction model requires only a single microphone to remove noise and output clean human voices, significantly enhancing the call quality of Bluetooth headsets.

「Translated content is generated by ChatGPT and is for reference only. Translation date:2024-05-19」

Recommend Cases

【解決方案】運用極現科技4D無人機雲端平台 巡檢成本降為五分之一
Utilizing Extreme Present Tech's 4D Drone Cloud Platform Reduces Inspection Costs to One-Fifth

The use of drones for intelligent inspection is becoming increasingly common, with major petrochemical and solar power plants continuing to adopt drone applications Located in Hsinchu, Extreme Present Technology earthbook has established a 4D cloud platform using its proprietary technology, offering drone, software, and data analysis platform services for intelligent inspections at solar power and petrochemical plants, reducing the total cost to just one-fifth of traditional methods involving hardware and software purchases, and cutting down the time from one month to approximately 24 hours, making it highly cost-effective For petrochemical industry operators who are constantly in a high-temperature, high-pressure dangerous environment, the safety control and inspection of plant facilities are critical 'As long as we can enhance the capabilities of facility inspection and risk identification in petrochemical sites, resource input is absolutely not an issue,' said a petrochemical industry representative with emphasis By implementing the drone 4D AI inspection cloud platform, the efficiency and safety of facility inspections among petrochemical operators can be elevated, further reducing the risk of equipment downtime Founded in March 2018, Extreme Present Tech has become a consistent winner in domestic entrepreneurship competitions, including being crowned champion in the 2019 OPEN DATA Business Innovation Practice, selected into Microsoft's startup accelerator in 2020, chosen for NVIDIA's AI startup team in 2021, and its products have been launched on the Microsoft Azure platform, earning investments from the National Development Fund and major domestic groups, thereby securing strong market validation for its technical prowess and services The founder and CEO of Extreme Present Tech, Hsu Wei-Cheng, mentioned that at the beginning of its establishment, the company took on the national space center's satellite 3D photography scheduling system and specialized in the integration of geographic information into 3D images As drone hardware technologies matured, the company shifted its operations towards the drone market and combined it with AI image recognition systems to establish a 4D cloud DaaS platform, offering services including online aerial photography ordering DaaS, 5GAIoT cloud platform SaaS, and enterpriseAPI server software, to meet the demands of drones in smart cities, facility inspection, engineering management, disaster response, pollution monitoring, and other applications, maximizing the value of drone services Smart aerial inspection regularly tracks the health status of plant equipment at a glance The quantity and area of petrochemical plants in Taiwan are immense, lacking sufficient manpower for comprehensive equipment inspections Given that petrochemical plants produce high-temperature flammable and corrosive chemicals that must be transmitted and stored through pipelines and tanks, long-term risks like pipeline ruptures and tank blockages could lead to severe occupational safety disasters, equipment downtime, and production stagnation Given the shortfalls in personnel for equipment inspections among petrochemical operators, Extreme Tech has already implemented a 4D AI drone inspection cloud platform combined with AI image recognition technology in petrochemical plant areas, providing ground-breaking evidence through the use of drones and proprietary app software services that connect on-site aerial data collection to the cloud platform, achieving fully automated and real-time aerial monitoring of petrochemical plant equipment pipelines, tanks, and ensuring precise locations and angles for each aerial operation, effectively compensating for the discrepancy in human inspection Hsu Wei-Cheng pointed out that the inspection drones used in petrochemical plants are equipped with dual lenses, one visible light and the other thermal infrared, which allow for determining pipeline obstructions through temperature conditions, enabling clients to immediately view the inspection status of the plant area from remote locations via the earthbook website, enhancing clients' inspection efficiency and accuracy The 4D aerial data platform meets diverse applications such as smart cities, transportation, engineering management, and pollution monitoring DaaS Online Order-Use Model Innovates Aerial Photography Business Model Saving 15 Costs Apart from providing a 4D aerial data platform, Extreme Present Tech also offers DaaS Drone as a Service After customers place orders on the website, Extreme Present coordinates with professionally licensed aerial photographers to provide on-site services Customers can monitor real-time operations through the platform and quickly obtain aerial data to evaluate any abnormalities, enabling timely alerts Take the solar power plant monitoring service as an example Given that solar power plant areas are large and widely distributed, located in the remote Pingtung area with the headquarters in Taipei, for inspections of the Pingtung plant, the customer just needs to use the DaaS service model, directly order online and upload a map of the Pingtung plant, obtain a quote from the company, and then entrust local Pingtung pilots to perform aerial inspections of the solar power plant During the process, the drone's route is automatically calculated by AI to plan the flight path, and the aerial data is transmitted to the client's cloud account, allowing the Taipei headquarters clients to immediately see the inspection status of the solar power plant from the earthbook website such as the condition of the solar panels, dust detection, or abnormal heat generation from solar electromagnetism, effectively helping the customer significantly reduce operational costs and efficiently complete the solar power plant inspection service Introduction of DaaS online aerial photography service in petrochemical plants According to estimates, solar power plant clients often incur high personnel costs by purchasing drones or outsourcing aerial photography With the long-term provision of aerial photography devices and the DaaS business model by Extreme Present Tech, customers can save 45 of aerial photography costs, and obtain aerial inspection reports within 24 hours post-operation, helping clients efficiently identify issues with solar panels Aiming to become the largest aerial data service company and enter the Southeast Asian market Since its establishment in 2018, Extreme Present Tech has rapidly grown in the aerial photography market with innovative thinking, actively expanding its aerial data application services Currently focused on cultivating the Taiwan market, the company aims to enter Southeast Asian nations, with Indonesia chosen as the first stop due to its high demand for infrastructure Hsu Wei-Cheng hopes that earthbook becomes the world's largest aerial data service platform Besides completing the initial round of funding from the National Development Fund and major groups, to penetrate the international market, the company continuously improves its drone data services and AI technology innovations, while also requiring the assistance of entities like the Industrial Technology Research Institute to find strategic investors that complement the company, fulfilling its goal of becoming an international aerial data corporation in phases Founder and CEO of Extreme Present Tech, Hsu Wei-Cheng「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」

這是一張圖片。 This is a picture.
AI Defect Intelligent Detection - Energy Reduction Smart Monitoring Solutions

AIIntelligent Defect Detection-Smart Monitoring Solution to Reduce Process Energy Consumption When there are over2ten thousand chip resistors on a ceramic substrate, how should one quickly detect defects The answer isUsingAIto detect。 In the era of rapid technological development, Leike proudly announces significant advances in its laser processing technology, thanks to the innovative applications of artificial intelligenceAILeike is committed to integrating advancedAItechnology into laser processing machines, and in2019year, in collaboration with partners, developed the world's first laser machining system that integratesAItechnology, and on this basis further developed in2023year the first ceramic substrate inspection machine that integratesAOIAILASERtechnology Smart Ceramic Substrate Inspection Machine Through the introduction ofAIand machine learning, along with the accumulation of big data samples, the system becomes smarter, which has led to improved product yield within one year5dramatically reducing the inspection time from originally2minutesper piece to just20secondsper piece, drastically lowering inspection costs, enabling efficient initial detection and post-laser marking to reduce waste in subsequent processes, diminishing overall carbon emissions of the site, allowing the automatic generation of detailed inspection reports for data analysis and optimization, which helps increase equipment capacity, reduce human error, enhancing the value of Leike's equipment, and strengthening the international competitiveness of the country's electromechanical industry Leike CorporationLaser TekFounded in1988year, and officially listed as a publicly traded company in2002year Since its establishment, it has become a leading global service provider and manufacturer of electronic packaging materials,SMDElectronic Packaging Materials,SMTinspection equipment, and laser systems Leike's general manager, with years of laser integration experience, observed that passive component customers can produce over20With many years of laser integration experience, he observed that the production capacity of passive component customers can exceed10billionSMDcomponents every month, but withSMDcomponents per month However, as component sizes continue to miniaturize, defect detection during production becomes increasingly challenging With thousands to millions of components on a single ceramic substrate, and as component sizes decrease and their laser processing positions become smaller, the difficulty of detection increases, making production inspection a critical process R-SMD Production Inspection Process AOIproblems of yield overkill relying onAIfor oversight, Yet,AOIthe inspection machine is a widespread and mature type, but the high accuracy on the marketAOIuses a technique that captures small images in a single shot and stitches them into a larger image Although accurate, this method requires more time for small-sizedSMDcomponents, which are more likely to be influenced by environmental factors like lighting and vibration that can cause misjudgments as a result,AOIyield rate can only be estimated by sampling, and components with poor sampling yield are not removed individually but discarded together with good ones manual re-inspection not only increases costs, but the lack of unified inspection standards ultimately results in about2-5products that are not detected as defective enter the subsequent manufacturing process monthly at least2,000thousands of such defective componentsSMDthat were not initially detected causing ongoing printing and machining inspections in subsequent processes Regardless of the waste of ink materials and energy, which increases the cost burden, this also accelerates equipment wear and shortens operational life Each stage of waste increases the site's carbon emissions, unfavorably impacting the company's carbon footprint Post-Adjustment Sample Photo Example 0402 TraditionalAOI High false positive rates in Automatic Optical Inspection AOI are a major production issue for manufacturers, particularly in the passive components industry where 'it's better to mistakenly reject a hundred than miss one'—a high standard, often leading to AOI setting extremely high parameters which makes devices overly sensitive Excessive stringency in data parameter settings can lead to high false positive rates For instance, if the dirt contamination on passive components resembles the color of the printing layers,AOI the misjudgment rate could reach 7 percent Contamination Dirt and Print Layer Color SimilarityAOIProne to Misjudgment Raytek stands apart from otherAOIsuppliers by discarding the stitching of small images or line scanning, effectively preventing data loss and discrepancies caused by hardware or environmental conditions during image processing It employs a large-array photodetector coupled with custom high-resolution lenses, using specialized imaging for composite processing Throughout this process, each pixel of the photodetector contains light information captured from various positions By combining this data, the image resolution and detail are enhanced, reaching a resolution of millions, and with multiple automatic light adjustments, a single shot can manage7070mmachieving an image resolution up to5umobtaining clear images, then throughSmart-AItechniques for analysis and selection Three Innovative Methods to Achieve Rapid InspectionSmart -AI Raytek's General Manager shares, rapidly implementingAItechnology and reducing inspection computation time, further developingSmart-AIthree major approaches Method one, initially useAOIto quickly separate good products from those with controversial defects, focusing the detection on the minority of defective identifications Method two, an automated labeling platform simplifies the training issue by using cameras to collect data from machines, automatic labeling replaces manual labeling, progressively training to improve accuracy The simpler the problem, the less data needed for training Method three,AOIandAIDual-track Advancement In the smart manufacturing process, relying solely onAOIorAIis not enough to accomplish the task alone, it must be preceded byAOIfirst marking the characteristics, distinguishing between good and defective parts, then usingAIa method for labeling and training Subsequently, by utilizing a repeating cascade effect, the detection benefits are greater as more training data accumulates,AOIreducing the ratio of errors,AIand gradually increasing the accuracy ratio Post Adjustment Object Detection and Training Through three major methods gradually building system reliability, and categorizing data for defect sorting, ultimatelyAIreturning the judgement results to the main system, utilizing laser machining to control truly defective products at the front end of the process, reducing the inflow of defective products into other stations, thus minimizing losses due to repeated tests or reprocessing Leading in smart laser equipment, chooseLASERTEKthe right one Continuously developed by the Taiwanese brand Raytek, combiningAIsmart detection and laser processing equipment to progressively build a smart monitoring solution stack from raw materials, products, testing, laser equipment, etc, aiming at reducing the energy consumption of the production process, implementing semiconductor advancements, substrates and component processing among other fields, producing equipment products capable of meeting the end-user demands under low carbon conditions, rapidly and with quality products and services expanding both domestic and international markets, enhancing the global competitiveness of localMade in TaiwanMITequipment 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-12-12」

【解決方案】瑕疵辨識率達百分百 耐銳利科技獲面板大廠青睞
Defect identification rate reaches 100%, Nairi Technology is favored by major panel manufacturers

On the machine tool production line, there are some slight differences in the first step of assembly Accumulated tolerances will cause the assembly work to be repeated, which is time-consuming and labor-intensive, resulting in shipment delays that will impact the company's reputation Narili Technology Company focuses on the field of smart manufacturing and provides various AI solutions It uses machine learning models to inherit the experience of old masters In the CNC processing machine assembly and casting process, it uses AI to analyze production line data, accurately adjust various data, and improve Production accuracy is 25 This AI production line data analysis system is called "Master 40" by Huang Changding, chairman of Naruili Technology It is the most evolved version of the master plus artificial intelligence It has been used in machine tool processing factories with remarkable results In addition, Nairi Technology used AI defect detection technology to participate in the 2021 AI Rookie Selection Competition of the Industrial Bureau of the Ministry of Economic Affairs, assisting AUO in advanced panel image defect detection, with an accuracy rate of 100, and won the award Assisted panel manufacturer AUO to solve problems with 100 accuracy in defect detectionHuang Changding further explained that during the production of general panels, edges and corners are There may be defects in the corners Although the defects are visible to the naked eye, AOI is often difficult to identify, causing the detection error rate to often exceed 30 Therefore, re-inspection must be carried out with manpower to improve the accuracy rate However, in response to the demand for a small number of diverse products and insufficient manpower, using AI detection is indeed a good method Nairui Technology, founded in 2018, has been able to win the favor of major panel manufacturers with its AI technology in just three years In fact, it has been honed in the field of CNC machine tools for a long time Tang Guowei, general manager of Narili Technology, pointed out that the top three CNC machine tool factories in Taiwan hope to introduce AI into the two production lines of assembly and casting Among them, on the assembly line, in order to maintain the accuracy of assembly, every part of the component is designed Tolerances are designed During assembly, each component is within the tolerance However, the cumulative tolerance still fails the final quality inspection and must be dismantled and reassembled This is not only time-consuming and labor-intensive, but also causes waste "After entering the production line, I realized that some masters have accumulated a lot of experience and are good at adjustment After his adjustment, the accuracy rate has improved a lot and the speed is faster" On the contrary, the new engineers did not Based on experience, it takes a long time to adjust and may not pass the quality inspection The yield rate of Master 40 system has increased significantly from 70 to 95Tang Guowei then said that the original size data set by Master during assembly All were recorded on paper After the information was written, it was stored in the warehouse and sealed No one studied the relationship between the dimensions Narili assists customers in designing the Fu 40 system Through the human-machine panel, the master can directly input the measured dimensions and related data during assembly After collecting data from different masters, AI algorithms are used to analyze the relationship between the data and create an AI model The AI model automatically notifies the operator what size to adjust to, and the quality inspection will definitely pass In this way, the yield rate will be improved It has increased significantly from 70 to more than 95 Narili Technology Company focuses on the field of smart manufacturing and provides various AI solutionsTang Guowei added, assembling the spindle of a CNC processing machine It took four hours In the first step, the machine made measurement errors, including vibration, temperature, speed, etc that were out of range It had to be dismantled and reinstalled, which took another four hours How to adjust after disassembly depends on the experience of the master At first, the master may have done the best assembly method based on experience, but the error rate was also 30, and the assembly took several days With the assistance of AI masters, the assembly time only takes half a day, and the yield rate reaches over 95, saving a lot of time and manpower "Use the AI model of machine learning to collect the experience of all the masters and provide it for AI learning The first step is digitalization, and the second step is knowledgeization This is the transformation of the enterprise "An important key", Huang Changding believes that Narili Technology is an important partner in the transformation of traditional manufacturing from automated production to digital transformation In addition, another industry that Naili Technology focuses on is the smart car dispatching system of the leading brand of elevator manufacturers The so-called car dispatch referring to the elevator car means that if there are more than two elevators, group management is required In the past, car dispatching was based on fixed rules If the elevator was closer to the requested car, that elevator would be automatically dispatched On the one hand, it did not take into account that dispatching a car if the elevator was called too many times might make other people wait longer The previous vehicle dispatching model did not take into account the usage characteristics of the building, resulting in a lot of waste For example, in an office building, there are peak hours in the morning, lunch break, and afternoon after work AI smart car dispatch can be flexibly adjusted according to off-peak and peak hours, increasing the efficiency of car dispatch, reducing waiting time, and reducing wasted electricity Introducing elevator smart dispatch to improve transportation efficiency and have environmental protection functionsHuang Changding added that just like the previous traffic lights at intersections, the system has already The number of seconds to stop and pass on highways, sub-trunks and small streets is programmed Smart traffic lights are now used to flexibly adjust waiting times to make road sections prone to congestion smoother Using AI to learn usage scenarios and introducing a smart dispatch system into elevators will improve transportation efficiency and make it more environmentally friendly In addition to introducing smart elevator dispatching, Nairili also introduced AI into the smart production and shipment scheduling system of elevator factories Elevator factories often cannot accurately estimate the customer's elevator delivery date For example, office buildings or stores must be completed to a certain extent before the elevator can be installed on the construction site If affected by unexpected factors such as delays in the customer's construction period, the elevator factory will often be idle or the schedule will be difficult to arrange Tang Guowei pointed out that generally those who understand the progress of client projects may be from business or engineering, but overall, the accuracy rate of shipments is only about 60, which means that 40 of them will not be shipped as scheduled Therefore, if the shipping schedule can be accurately estimated, the production line can be freed up for emergency orders or other product production needs The AI smart scheduling system will analyze past shipment data, about 20-30 parameters such as climate, distance between the factory and the construction site, and customer credit, and put them into the AI algorithm to accurately predict whether shipments can be made as scheduled goods Huang Changding also specifically stated that the machine learning of Naili Technology is not ordinary machine learning, but also incorporates various calculation methods such as traditional image processing technology and statistics Only by being very familiar with the domain knowledge can we make good products AI models are also where the company’s competitiveness lies He emphasized that the data that general SaaS platforms can process is very limited, and the accuracy rate has increased from 70 to 75 at most Naili’s strength lies in AI algorithms and machine learning, and it must be coupled with in-depth industry knowledge to produce output Good AI model Narili Technology started with the AI project, gradually deepened the technology, chose to start with the more difficult tasks, and accumulated rules of thumb It is expected to develop SaaS services this year 2022, based on customer needs starting point, gradually gaining a foothold and becoming an important partner in smart manufacturing The picture left shows the general manager of Naruili Technology Tang Guowei and Chairman Huang Changding right「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」