Newsroom

<figure class="image"><img style="aspect-ratio:800/524;" src="https://deeply-website-backend.s3.ap-northeast-2.amazonaws.com/deeply/2024-11/598bf2e5-d577-4e87-bdf8-4ab7aa8ad07c.jpg" width="800" height="524"></figure>▲ (From left) Suji Lee, CEO of Deeply, and Myunghoon Ryu, CIOInterview with Suji Lee, CEO of DeeplyAs AI technology moves beyond visual recognition into auditory analysis, it becomes increasingly integrated into daily life and work environments. Deeply, an AI startup focused on sound-based solutions aims to enhance efficiency and create new value across various industries. We spoke with Suji Lee, CEO of Deeply, about the company’s founding, technological evolution, and long-term plans.&nbsp;Preventing Risks with 'ListenAI'Deeply, it has gained attention for developing AI solutions that use sound data for emergency detection and industrial applications. Their flagship product, ListenAI, analyzes real-time patterns such as screams, collisions, and machinery noises to address safety issues in industrial and public spaces. With a focus on B2B solutions, Deeply aims to implement predictive maintenance systems that boost efficiency and safety in manufacturing.&nbsp;What sets ListenAI apart is its ability to analyze complex audio environments and recognize patterns to make real-time assessments, going beyond just detecting volume or vibrations. This capability has been deployed in subways, large shopping centers, and factories, offering precision previously lacking in AI systems. This precision improves industrial safety and enables predictive maintenance, strengthening Deeply's competitive advantage.&nbsp;ListenAI has demonstrated its capabilities through deployments at various public and private sites.. Notable examples include installations at the Government Sejong Complex, Kangwon Land, and Incheon National University Station. At the Government Sejong Complex Sports Center, over 10 units of ListenAI were installed to monitor blind spots that CCTV cannot cover, ensuring comprehensive emergency detection. At Kangwon Land, the system detects and preempts conflicts between patrons, allowing for immediate intervention. In Incheon, ListenAI identifies sounds such as female screams and violent altercations, promptly alerting control centers. “The system can analyze and trigger warnings within seconds of sound detection, helping prevent real-life threats,” said CEO Lee.&nbsp;Predictive Maintenance in ManufacturingDeeply’s sound analysis technology is also making waves in the manufacturing sector through its predictive maintenance applications. Their solution detects potential malfunctions in rotating machinery and motors by analyzing abnormal sound patterns. For instance, ListenAI can identify vibration patterns indicative of potential bearing failures, issuing warnings before breakdowns occur. This approach has earned high praise from manufacturers, outperforming traditional vibration-based monitoring systems in both speed and precision.&nbsp;“Minor issues in machinery that go undetected can lead to significant accidents. However, current vibration sensors can’t detect all internal noises or abnormal behaviors. Our AI solution comprehensively analyzes sound signals to precisely diagnose machine status,” explained Lee. This technology helps plant operators avoid unexpected equipment failures, reducing downtime and maintenance costs.&nbsp;“Establishing a Robust Safety Framework with High-Quality Data”Deeply’s AI solutions prioritize contextual understanding of sound, built on a vast dataset exceeding 50,000 hours. This extensive collection includes not only conventional sound data but also situation-specific and global environmental audio, ensuring high accuracy and reliability. The company’s strong dataset foundation is one of its key strengths. Early in its journey, Deeply enhanced its AI capabilities with sound data collected from home environments during its initial B2C phase.&nbsp;As it shifted to B2B, Deeply expanded its data collection efforts internationally, including field data from locations like India. Gathering audio from crowded streets with varied noises such as car horns and emergency alerts, Deeply developed models capable of accurately detecting emergencies in complex global environments. This expertise has positioned Deeply in voice recognition and as a comprehensive solution for analyzing and classifying all forms of unstructured sound data.&nbsp;Today, Deeply offers tailored solutions across industries, customizing models to meet client needs. For example, in manufacturing, the system collects and analyzes data from on-site microphones and servers to detect anomalies and assist with prompt interventions. Integrating with traditional vibration and ultrasonic sensors enhances predictive accuracy, and Deeply’s AI continuously learns to improve its precision. In the safety sector, ListenAI works in tandem with CCTV systems, combining visual and audio data for greater reliability, showcasing Deeply’s multimodal approach.&nbsp;Looking ahead, Deeply aims to further leverage its sound recognition technology to enhance safety and productivity across industries. “AI has evolved from a supporting tool to one that supplements human senses, like hearing,” said CEO Lee. “We’re advancing toward an AI that can understand and respond to complex sound environments autonomously.” To achieve this, Deeply plans to expand its datasets, develop new algorithms, and strengthen partnerships to provide customized solutions for various sectors.&nbsp;Deeply is also developing an integrated solution combining hardware and software to accelerate market entry. Beyond offering software, the company supplies custom microphones and servers to ensure optimal performance in diverse environments. For example, high-frequency and vibration-detecting microphones are used in manufacturing settings, while noise-filtering features are incorporated for noisy sites like construction areas or subways, enhancing accuracy.&nbsp;By providing tailored solutions for different industry needs, Deeply seeks to expand the applications of AI sound recognition technology. “Our ultimate goal is to go beyond just offering sound analysis AI solutions to becoming a comprehensive platform that revolutionizes safety and efficiency across all industries,” concluded CEO Lee.&nbsp;— Jaechang Seo, HelloT News &nbsp;<a href="https://www.hellot.net/news/article.html?no=9515">https://www.hellot.net/news/article.html?no=9515</a>&nbsp;

[HelloAI] Deeply Sets a New Standard in Safety with 'Sound Understanding AI'

관리자

<figure class="image"><img style="aspect-ratio:426/255;" src="https://deeply-website-backend.s3.ap-northeast-2.amazonaws.com/deeply/2024-11/23abc4d6-ee07-4ea0-a463-bb0d2ec2b1c1.jpg" width="426" height="255"></figure>Suji Lee (right), CEO of Deeply, and Myunghoon Ryu, CIO.&nbsp;Deeply, a startup specializing in AI sound detection and analysis solutions is making strides in the field. In an interview with The Korea Economic Daily, CEO Suji Lee emphasized, “The potential of sound AI is immense—from detecting abuse situations in call centers to monitoring livestock health on farms.”&nbsp;Deeply’s proprietary sound detection system, ListenAI, utilizes a dataset of 42 different sound types, totaling 50,000 hours of audio. This enables it to identify sudden incidents across various environments. For example, it can detect machine malfunctions in manufacturing facilities or alert workers to fall accidents at construction sites through sounds like crashing or shouting.While existing AI services focused on voice analysis for translation, specialized AI for sound classification and analysis has been rare. CIO Myunghoon Ryu noted, “We have developed an integrated solution combining AI software with specialized hardware like microphones and edge servers.”&nbsp;Deeply collaborates with major partners, including Kangwon Land, Lotte Construction, and KORAIL, to implement its solutions effectively. More recently, ListenAI has been implemented at facilities like the Government Sejong Complex Sports Center and restrooms at Incheon National University Station. The system detects sounds in real-time and issues alerts upon identifying abnormal situations. CEO Lee stated, “Even where CCTV is installed, incidents can be missed without continuous monitoring. ListenAI enhances the accuracy of surveillance when used alongside CCTV.” KORAIL also employs the solution to monitor the sounds of traction motors for early fault detection.&nbsp;Lee founded Deeply after identifying the technical challenges in sound AI compared to visual AI. He said, “For sound AI to function effectively, it needs to account for factors like echo, noise, and human speech characteristics. This makes it a field that is not easily replicable by other AI companies.”&nbsp;To develop their sound AI models, Deeply gathered diverse sound data, including screams, fighting noises, and object impacts. Ryu shared, “We acquired external data, hired actors for specific recordings, and curated a dataset tailored to sound event lists for training models.”&nbsp;Deeply has also embarked on a project with Korea University Hospital to track the daily routines of depression patients, monitoring activities such as regular exercise and meals through sound analysis.&nbsp;“The possibilities for sound AI are endless,” said Lee. “Deeply aims to be at the forefront of this pioneering field.”&nbsp;— Euni Ko, The Korea Economic Daily<a href="https://www.hankyung.com/article/2024111129051">https://www.hankyung.com/article/2024111129051</a>

“AI Sound Analysis: Endless Applications”

메인관리자

Effective ways to prevent accidents and crimes&nbsp;<figure class="image"><img style="aspect-ratio:600/298;" src="https://deeply-website-backend.s3.ap-northeast-2.amazonaws.com/deeply/2024-11/5d9d3b6f-9c81-40cf-8032-32c1b81a24bc.png" width="600" height="298"></figure> Listen to AI, an artificial intelligence solution that understands the meaning of sound.Capable of detecting over 100 distinct sounds while filtering out ambient noise to focus on target voicesDeeply, a technology startup, is researching solutions to analyze various sound sources and voices. It has developed ‘Listen AI’, a sound-based situation detection artificial intelligence solution that analyzes the shape of sound using deep learning and responds to various events and accidents. Listen AI detects various sound sources like the human ear and analyzes what the current sound means. It detects a wide range of sounds, including screams, moans, commotions, vehicle crashes, explosions, and gunshots, and connects it to CCTV and dash cameras with edge devices. This creates a system that can detect situations and send information to the respective center for quick response. Listen AI can detect and customize over 100 types of sound, and can quickly integrate with and respond to devices in the control center via an application programming interface (API) and software development kit (SDK). Listen AI is unique in its ability to filter out even the most meaningless ambient noise.&nbsp;<figure class="image"><img style="aspect-ratio:600/459;" src="https://deeply-website-backend.s3.ap-northeast-2.amazonaws.com/deeply/2024-11/6db6db34-ce12-4ff0-8f96-a6b56d058cff.png" width="600" height="459"></figure>Deeply CEO Lee Suji says “Unlike cameras, speech analytics has no blind spots, which is a big advantage, but it’s hard to filter out ambient noise. However, our solution can clearly distinguish between sounds coming from various media such as TVs and speakers, and emergency calls amidst the noise of everyday life. Because it can clearly distinguish between noise and target sounds, the false positive rate is very low and allows us to react immediately.”&nbsp;&nbsp;<figure class="image"><img style="aspect-ratio:600/209;" src="https://deeply-website-backend.s3.ap-northeast-2.amazonaws.com/deeply/2024-11/c939888d-c8c5-4f60-9576-6ecd923426ba.png" width="600" height="209"></figure>Boasting TTA authentication accuracyListen AI is TTA-certified, highly accurate and technologically advanced, with over 98% accuracy on F1 scores. Deeply has the world’s largest collection of nonverbal sound data, verified by more than 200,000 users worldwide, and includes 42 nonverbal sound datasets with more than 50,000 hours of audio. In addition to safety solutions, the innovative Listen AI provides machine solutions that detect abnormal sounds from machines or equipment to continuously monitor the entire process and equipment status. The innovative Listen AI system is currently deployed in 10 restrooms across the Incheon subway system, and in the Sejong Government Complex, it is also installed in places where CCTV cannot be installed, such as toilets. Deeply is also operating a system in collaboration with Korail that analyzes the sound of vehicles for malfunctions and abnormalities and predicts failures on high-speed rail, collecting all sounds generated when various parts malfunction and building them into big data. More recently, Deeply’s solution has also been applied to military drones, where it can be mounted on large unmanned drones and used by air traffic controllers to check for abnormalities. CEO Lee Suji published a paper at ICASSP 2024, an international conference on signal processing, in July of this year, and won the Director’s Award at the Digital Future Innovation Awards. 신태섭 기자&nbsp; <a href="mailto:tss79@naver.com">tss79@naver.com</a><a href="http://www.powerkoream.co.kr/news/articleView.html?idxno=1108798">http://www.powerkoream.co.kr/news/articleView.html?idxno=1108798</a>&nbsp;