Frequently asked questions

Deeply is an AI startup founded in 2017 that delivers sound analysis solutions across manufacturing, safety, and security use cases.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Sound AI automates the auditory inspection that used to rest on a person’s ears. A judgment made by ear or by fingertip shifts with the worker’s fatigue and leaves no inspection history behind as data, while Listen AI makes the same judgment from the sound signal and keeps every OK/NG result as a record. More than 60% cost reduction has been reported for large hearing inspection line operations. Accuracy and measurement range differ by inspection type, and the Listen AI Industrial page covers them per type in its own FAQ.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Listen AI judges from the sound signal rather than from an image or a vibration reading. It reaches the motor and actuator defects inside a product that a vision sensor misses, inspecting every unit 24/7 at the EOL stage and blocking 99.8% of defect escapes. That figure is scoped to full inspection of motors and actuators at the EOL stage and is not combined with figures from other inspection types. Sounds too faint for a person to tell apart, such as a connector engagement click, are judged the same way.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

The three core Listen AI Industrial solutions are rotating-equipment noise inspection, assembly-line connector engagement sound inspection, and equipment aging detection. On the assembly process it overcomes the high noise of an OEM assembly line and detects continuous connector and bolt fastening work in real time with 99%+ accuracy. For predictive maintenance, which watches the equipment rather than the product, it draws on a 15TB manufacturing acoustic dataset and analysis experience across more than 70 sound types to catch a fault signal early and stop downtime before it starts. Deployed cases also cover glass breakage and scratch detection, semiconductor component wear diagnosis, robot equipment aging analysis, and gas and pressure valve sound analysis.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

What it takes is the equipment to be inspected and the sound that equipment makes. Once the hardware is installed and the engagement or operating sounds of the target equipment are collected, initial evaluation can begin on top of Deeply’s Foundation Model. Within the scope of the connector engagement sound press release, this does not wait on weeks or months of site-specific defect data and changes no existing line equipment, tools, or layout, but that account assumes a Foundation Model specialized in manufacturing acoustic data, so it does not mean every deployment needs no data collection at all. Inspection history can be linked to a MES (Manufacturing Execution System) or PLC and kept as traceable data.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

A PoC runs in four steps: install the hardware, collect sound from the target equipment, run the initial evaluation, then carry the result into production-line deployment. The initial evaluation starts on top of Deeply’s Foundation Model, and the PoC result is what carries into the production line. The microphone setup and the analysis server deployment are chosen against the noise conditions on site, and the Listen AI Industrial page covers the conditions per inspection type in its own FAQ.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Yes. Listen AI Industrial is designed for environments above 85 dB where impact wrenches, air guns, metal friction, and conveyor belts overlap. Deeply’s proprietary Foundation Model is built on more than 2.1 million hours of real-world factory noise data and more than 10 million process event data points, which lets it separate the faint engagement sounds workers previously judged by ear or fingertip from the surrounding noise.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

The three core Listen AI Industrial solutions are rotating-equipment noise inspection, assembly-line connector engagement sound inspection, and equipment aging detection. Deployed cases also cover glass breakage and scratch detection, semiconductor component wear diagnosis, robot equipment aging analysis, large engine drive-unit aging analysis, and gas and pressure valve sound analysis.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Listen AI Industrial reports 99.87% accuracy on assembly-line connector engagement sound inspection, measured on global automaker Company H production lines in Korea and Mexico and published in a June 11, 2026 press release. Rotating-equipment noise inspection is separately reported at 99.78% accuracy with sub-second inspection time in earlier industrial coverage. The two figures cover different inspection types and measurement scopes, so they are not combined into a single claim.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

It decides through Frequency-band Analysis. Listen AI Industrial distinguishes the Primary Lock in the 3–16 kHz band and the Secondary Lock in the 7–16 kHz band, then returns an OK or NG result. This is what catches half-clicks and soft connections that never fully engaged.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Yes. Listen AI Industrial inspects in about one second per product to run in-line 100% inspection, judging 6,000 products an hour and 4.3 million a month fully in-line. At the EOL stage it inspects every motor and actuator 24/7. Inspection time and judgement volume shift with the inspection type and the process conditions, so whether a given line’s Cycle Time is met is confirmed during the PoC.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Yes. Listen AI Industrial links inspection history to a MES (Manufacturing Execution System) or PLC so results are retained as traceable data. Deployment is supported as an edge analysis server setup or on-premise.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

The setup pairs microphones chosen for the target sound and the site with an analysis server. Microphones are selected from fixed directional microphones, wristband microphones worn by workers, and array microphones for high-noise areas. The analysis server is supported as an edge analysis server setup or on-premise, and OK/NG results are linked in real time with an existing MES (Manufacturing Execution System) or PLC so they are retained as traceable data. In the connector engagement press release, installation is described as requiring no change to the existing line’s equipment, tools, or layout.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Yes. Rather than waiting weeks or months while site-specific defect data is collected, Listen AI Industrial begins initial evaluation on Deeply’s Foundation Model from the moment the hardware is installed. Collecting the engagement or operating sounds of the target equipment is enough to evaluate on top of it. This is scoped to the connector engagement press release and its manufacturing-specific Foundation Model, so it does not mean every deployment can proceed without data collection. PoC results then carry into production-line deployment.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

They judge different things at different moments. Quality inspection judges the product being made: on the production line it detects continuous connector and bolt engagement work in real time, and at the EOL stage it inspects every motor and actuator before shipment to block defect escapes. Predictive maintenance (PdM) judges the equipment that makes the product rather than the product itself. Its purpose is to catch equipment anomaly signals such as bearings or gas leak sound early enough to prevent downtime, and it judges from continuous monitoring of the equipment rather than at one point in a process. Because the two cover different measurement scopes, accuracy figures are not carried from one to the other.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Yes. Listen AI Safety recognises danger by sound rather than video, which makes it suited to restrooms, changing rooms, saunas, and other spaces where cameras cannot be installed. Live deployments such as the Government Buildings Management Office gymnasium and the public restrooms of Naejangsan National Park centre on exactly these spaces.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Listen AI Safety detects 12 sounds, among them screams, aggressive yelling, crying and sobbing, heavy breathing, breaking glass, physical impacts, sirens, and car horns. Which sounds are detected, and at what threshold, is set per microphone for each site.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

The purpose of Listen AI Safety is to detect the kinds of sounds that indicate danger. Sound is analysed inside the on-site AI SOUND BOX and discarded as soon as analysis is complete; only the detected sound type reaches the customer system, never the recording itself. Microphone input and event delivery run over physically separate network interfaces, and a legal review against Article 14 of the Korean Protection of Communications Secrets Act has been completed.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Through composite events. Conditions set for each site combine several detections, such as a scream, aggressive yelling, and an impact within a set number of seconds, into one danger situation, so one-off sounds no longer raise unnecessary alerts. Per-microphone thresholds and the ability for operators to listen and judge for themselves are provided as well.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Alerts appear as real-time pop-ups in the monitoring system. The Open API and the ONVIF and MQTT protocols connect Listen AI Safety to existing VMS and control-room systems and to emergency call buttons, and an embedded SDK for hardware integration is available by arrangement.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

Listen AI Safety is a designated Innovative Product of the Korean Public Procurement Service, so public institutions can purchase it through the KONEPS national procurement system. Full product details and purchasing terms are available on the Public Procurement Service innovation marketplace.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.

The danger-sound detection performance of Listen AI Safety was verified at an F1-score of 98% or higher in TTA (Telecommunications Technology Association) test certification. Results vary with the noise conditions of each site, so configuration and validation for the site are carried out before rollout.

Explore this answer with AI

Choose an AI to explore this with.

ChatGPTOpens in a new windowClaudeOpens in a new windowPerplexityOpens in a new windowGeminiOpens in a new window

Copy question

If the question is not filled in, copy and paste it.