Real-World Applications
mpWAV provides the full voice interface stack — single-microphone noise control, multi-microphone echo and noise preprocessing, wake-word detection, sound source localization, speaker diarization, on-device speech recognition, and a conversational LLM.
For product environments as different as robots, kiosks, vehicles, smart devices, and meeting systems, we select the required technologies and connect them into one working architecture.
Key applications
Robots · Kiosks · Vehicles & Mobility · Smart Devices & Earbuds · Meetings & Voice Chat · Factory Anomaly Detection · Defense · Hearing Assistance
Robot · Kiosk · Mobility · Smart Device · Meeting · Factory · Defense · Hearing Support
↓ mpWAV Voice Interface Stack
↓ Listen · Understand · Respond
Technology Matched to the Environment
mpWAV analyzes your product's microphone count, speaker output, user position, number of speakers, dominant noise, and service goal to compose the right combination of voice technologies.
Earbuds with one microphone start from mpNC; multi-channel robots from mpAEC, mpBeamforming, and mpLocalization; meetings from mpS and mpDiarization; conversational kiosks from mpASR and mpLLM.
We don't force one generic algorithm onto every product — we design the voice stack around the product's conditions.
Single Mic → mpNC | Multi-Channel+Noise → mpBeamforming | Speaker+Mic → mpAEC/mpAB
Wake Word·Direction → mpWWD/mpLocalization | Multiple Speakers → mpS/mpDiarization
Recognition → mpASR | Dialogue·Intent → mpLLM
Application Overview
Multi-channel voice interfaces for products that move and speak
Helps robots detect the user's call, find the direction the voice came from, and recognize commands amid their own speakers and motor noise.
mpWWD · mpLocalization · mpAEC · mpBeamforming · mpAB · mpASR · mpLLM
See robot applicationsVoice ordering and on-device dialogue in store noise
Captures the user's voice with a multi-channel array, reduces store noise and prompt echo, then understands order intent and options through recognition and a conversational language model.
mpAB · mpASR · mpLLM · Mic-Array HW
See kiosk applicationsVoice commands amid driving noise and car audio
Reduces car audio echo and road noise, then analyzes the wake word, seat or speech direction, and voice commands.
mpWWD · mpLocalization · mpAEC · mpBeamforming · mpAB · mpASR · mpLLM
See vehicle applicationsSingle-microphone processing for products that can't add mics
For earbuds, wearables, and small smart devices where an array isn't possible, one microphone is enough to reduce the impact of ambient noise.
mpNC · mpWWD · mpASR · mpLLM
See smart device applicationsSeparating voices and organizing them by speaker
Prepares multi-party and overlapping speech for downstream processing, tells who spoke when, then connects meetings to transcripts and summaries.
mpS · mpDiarization · mpAEC · mpBeamforming · mpASR · mpLLM
See meeting applicationsDetecting abnormal signs amid complex equipment noise
Analyzes production line noise and implements acoustic analysis that distinguishes normal from abnormal motor and equipment sounds.
Noise-Robust Acoustic Processing · Anomaly Detection · Edge Integration
See factory applicationsVoice interfaces for severely noisy environments
Voice preprocessing, wake-word detection, on-device recognition, and dedicated hardware structures for strong ambient noise and constrained networks.
mpAEC · mpBeamforming · mpAB · mpWWD · mpASR · HW · SoC
See defense applicationsClearer everyday conversation in noise
Through ClearSense Audio on a smartphone with ordinary earphones, mpWAV voice enhancement extends to everyday conversation and public hearing welfare.
mpNC · Speech Enhancement · ClearSense Audio
See hearing assistanceService robot | Voice-order kiosk | Car cabin | Earbuds·wearables
Meeting room·voice chat | Factory equipment | High-noise environments | ClearSense in use
Robotics
Robots run motors and cooling fans while delivering prompts and responses through their own speakers.
Users may call from any direction, not just the front, while nearby conversation and room reflections also enter the microphones.
mpWAV connects wake-word detection, localization, echo cancellation, multi-microphone noise reduction, and recognition-plus-dialogue into the robot's complete voice interface.
mpWWD
Detects the wake word and activates the robot's listening state.
mpLocalization
Estimates the direction or position of the user's voice.
mpAEC
Reduces echo from the robot's own prompts and responses.
mpBeamforming·mpAB
Reduces motor, fan, and nearby-conversation noise, strengthening the target voice.
mpASR
Converts commands into text or robot control commands.
mpLLM
Interprets natural-language commands, questions, and dialogue context.
Example pipeline
Wake Word → mpWWD
→ User Direction → mpLocalization
→ Mic Array + Speaker Reference → mpAB
→ mpASR → mpLLM / Robot Command
Official material presents speech recognition preprocessing validation in home robot, showroom robot, and care robot environments.
Wake word · user direction · motor/fan noise · robot speaker · multi-channel mics
mpWWD → mpLocalization → mpAB → mpASR → mpLLM
Conversational Kiosk
Accurate recognition alone is not enough for kiosk voice ordering.
Users phrase menus and options in many ways, missing information requires follow-up questions, and the final result must reach the ordering system.
mpWAV connects a linear microphone array, mpAB preprocessing, mpASR recognition, and mpLLM — a 1B-class on-device dialogue model direction — into a conversational ordering pipeline.
Linear mic array
Captures far-field user speech in front of the kiosk on multiple channels.
mpAB
Processes store conversation, background music, and kiosk prompt echo together.
mpASR
Converts menu names, options, quantities, and order utterances into text.
mpLLM
Interprets intent and context, asking follow-up questions for missing information.
Example pipeline
User Speech → Linear Mic Array → mpAB
→ mpASR → mpLLM
→ Menu · Option · Quantity → Ordering API
Example conversation
Official material presents deployment experience with multi-channel microphone I/O and ASR preprocessing modules for kiosks.
Mic Array → mpAB → mpASR → mpLLM → Ordering API
Automotive and Mobility
A car cabin combines road noise, wind, HVAC, music, and passenger conversation at once.
The voice interface must recognize the wake word, identify which seat or direction spoke, cancel car audio echo, and then understand commands and natural-language requests.
mpWWD
Detects the automotive wake word.
mpLocalization
Estimates speech direction and seat position.
mpAEC
Reduces echo from car audio and voice prompts.
mpBeamforming·mpAB
Reduces driving noise and passenger conversation.
mpASR
Recognizes navigation, climate, and media commands.
mpLLM
Connects natural-language requests and context to vehicle functions.
Example pipeline
Wake Word → Seat/Direction Detection
→ Cabin Mic Array + Audio Reference → mpAB
→ mpASR → mpLLM / Vehicle Intent
→ Infotainment · Navigation · Control
Official material describes reviewing noise robustness and global expandability in the mobility field.
Driver·passengers · cabin mic array · audio speakers · road/wind noise
Direction estimation · mpAEC·Beamforming processing
Smart Devices and Earbuds
Earbuds, wearables, and small smart devices often cannot fit a multi-microphone array due to size, layout, and power constraints.
mpNC reduces the impact of ambient noise from a single microphone input, making voice input improvement and a voice interface feasible on single-mic products.
mpNC
Reduces everyday ambient noise from a single microphone input.
mpWWD
Detects the wake word to activate product functions.
mpASR
Recognizes commands and short utterances on-device.
mpLLM
Connects product functions with natural-language requests.
Example pipeline
Single Microphone → mpNC → mpWWD → mpASR → mpLLM → Device Function
Applicable products
Products that cannot fit a multi-channel array are covered by mpNC-based single-microphone processing.
Single Mic → mpNC → Wake Word → ASR → Device Action
Meeting and Voice Chat
Meetings and voice chat can combine multiple talkers, speaker echo, far-field voices, and overlapping conversation.
mpWAV uses mpS to prepare multi-party input for downstream processing, mpDiarization to tell who spoke when, then mpASR and mpLLM to extend into transcripts, summaries, and Q&A.
mpS
Separates and organizes multi-party or overlapping input for downstream processing.
mpDiarization
Separates per-speaker segments and speaker turns.
mpAEC
Reduces echo from speakers or remote participants re-entering the microphone.
mpBeamforming
Reduces far-field voices and ambient noise in the meeting room.
mpASR
Transcribes the meeting audio.
mpLLM
Extends into summaries, key decisions, and Q&A.
Example pipeline
Multiple Speakers + Speaker Echo
→ mpAEC / Beamforming → mpS → mpDiarization
→ mpASR → mpLLM
→ Transcript · Speaker Tag · Summary
Applicable services
Mixed Meeting Audio → mpS → Speaker A/B/C → mpDiarization → mpASR → mpLLM Summary
Industrial Acoustic Anomaly Detection
On a production line, many motors and machines run at once, mixing normal and abnormal sounds.
Building on its real-environment noise processing, mpWAV organizes equipment acoustic signals and implements analysis that distinguishes normal from abnormal patterns.
Acoustic preprocessing
Extracts the target equipment's acoustic signature from complex line noise.
Anomaly model
Detects state changes by comparing normal and abnormal data.
Edge processing
Collects and analyzes acoustic data close to the equipment.
License·co-development
Jointly develops data-driven models fit to your equipment and production environment.
Example pipeline
Factory Acoustic Input → Noise-Robust Processing
→ Acoustic Feature Analysis → Normal / Anomaly Classification
→ Inspection · Monitoring · Alert
Applicable areas
Official material presents validating motor anomaly detection on a noisy production line without a separate test chamber. Detection accuracy, equipment types, and production status are not specified in public material.
Capture → Noise processing → Feature analysis → Normal/Anomaly → Alert
Defense and High-Noise Environments
Defense and special environments can require handling strong ambient noise, limited networks, and embedded compute constraints together.
mpWAV reviews technology collaboration combining multi-channel preprocessing, wake-word detection, on-device recognition, and FPGA·DSP·SoC structures.
mpAEC · mpBeamforming · mpAB
Multi-channel preprocessing amid strong ambient noise.
mpWWD
Wake-word based interface activation.
mpASR
Network-independent on-device speech recognition.
FPGA·DSP·HW · Dedicated SoC
Embedded real-time processing and dedicated chip structures.
Directions
High-noise voice commands · network-independent execution · multi-channel input
Embedded real-time processing · dedicated voice interface chips · per-product co-development
High-Noise Audio → mpAB → mpWWD → On-Device mpASR → Embedded Control
Hearing Support and Smart Listening
ClearSense Audio is a smart listening solution that helps people hear conversation in ambient noise more clearly, using a smartphone and ordinary earphones.
It extends mpWAV's single-microphone noise control and voice enhancement into B2C products and public hearing welfare.
mpNC·Voice enhancement
Reduces ambient noise in conversation captured by the microphone.
Mobile processing
Supports product structures using a smartphone and ordinary earphones.
User-centered interface
Everyday listening that is easier to access than specialist equipment.
Usage environments
Family conversation · restaurants & cafés · welfare center programs · public hearing welfare · everyday listening
Official material presents ClearSense Audio's welfare center validation and public support program experience.
Older user + smartphone + ordinary earphones — family/welfare center conversation
Application Technology Matrix
| Application | Input | Interaction | Recognition & Dialogue | Implementation |
|---|---|---|---|---|
| Robots | mpAEC·Beamforming·mpAB | mpWWD·Localization | mpASR·mpLLM | Mic Array·DSP·FPGA·AP |
| Kiosks | mpAB | mpWWD as needed | mpASR·mpLLM | Linear Mic Array·AP·DSP |
| Vehicles & mobility | mpAEC·Beamforming·mpAB | mpWWD·Localization | mpASR·mpLLM | DSP·AP·SoC |
| Smart devices & earbuds | mpNC | mpWWD | mpASR·mpLLM | Edge·AP·lightweight models |
| Meetings & voice chat | mpAEC·Beamforming·mpS | mpDiarization | mpASR·mpLLM | PC·Server·Edge |
| Factory anomaly detection | Acoustic Processing | Anomaly Analysis | Analysis models as needed | Edge·License |
| Defense & special environments | mpAEC·Beamforming·mpAB | mpWWD | mpASR | FPGA·DSP·SoC |
| Hearing assistance | mpNC·Speech Enhancement | — | — | Mobile·ClearSense |
This table is a general guide. Real projects finalize the combination based on product hardware, data, latency, and service goals.
NC AEC BF AB WWD S LOC DIA ASR LLM HW × Robot·Kiosk·Mobility·Smart·Meeting…
Find the Right Technology
| Product problem | Recommended technology |
|---|---|
| Only one microphone available | mpNC |
| Echo from the product's own speaker | mpAEC |
| Reducing ambient noise with multiple mics | mpBeamforming |
| Echo and noise at the same time | mpAB |
| Detecting a wake word | mpWWD |
| Separating multi-party voice input | mpS |
| Knowing the speaker's direction or position | mpLocalization |
| Telling who spoke when | mpDiarization |
| Converting speech to text or commands | mpASR |
| Understanding context and intent | mpLLM |
| Multi-channel audio and real-time processing | Multi-Channel HW |
Mics? 1 → mpNC / several → mpBeamforming
Speaker present → mpAEC/mpAB | Wake·direction·speakers → mpWWD/Localization/Diarization
Recognition·dialogue → mpASR/mpLLM
Delivery Options
Preprocessing and interaction technology applied in front of your product and ASR.
Before-and-after comparison with your real product audio and field data.
Multi-channel input, speaker output, and the AEC reference connected in one structure.
Runs mpWAV technology on your embedded compute platform.
Fits mpASR and mpLLM to the target product's memory and compute budget.
Preprocessing, acoustic analysis, and conversational interfaces co-developed around your product data.
Dedicated voice interface structures for volume production and miniaturization.
Software PoC → Mic/HW Evaluation → DSP·FPGA·AP Porting → Product Integration → License/Co-Dev → Semiconductor IP/SoC
Application Experience
ASR preprocessing for home, showroom, and care robots
Multi-channel microphone arrays and ASR preprocessing modules
Voice interfaces in real noise, with global expandability reviewed
Motor anomaly detection validated amid complex production noise
ClearSense Audio usability reviewed in welfare center environments
Robot voice test | Kiosk linear array | In-vehicle data | Factory motor measurement | ClearSense welfare center
From Use Case to Product
Wake-up, commands, dialogue, meeting records, or anomaly detection — we scope what you need.
Microphone count, speaker output, user distance, and dominant noise.
Single/multi-channel raw audio, the AEC reference, and your current ASR results.
Select from mpNC, mpAEC, mpBeamforming, mpAB, mpWWD, mpS, mpLocalization, mpDiarization, mpASR, and mpLLM.
Evaluate audio quality, detection rates, localization and diarization, recognition, dialogue results, and latency.
Apply as software, a mic array, DSP, FPGA, AP, or an edge environment.
Deliver recognition and dialogue results to robot control, ordering, vehicle functions, meeting systems, or alerting.
Verify performance where the product is actually used, under its real noise.
Review follow-up products, platforms, model optimization, and SoC expansion.
Scenario → Acoustic analysis → Data → Stack → PoC → Integration → API → Field → Production
FAQ
Robots, kiosks, vehicles, smart devices and earbuds, meetings and voice chat, factory acoustic anomaly detection, and hearing assistance.
Defense and special environments are handled as dedicated hardware and co-development areas.
Yes.
mpNC reduces ambient noise from a single microphone input — the technical foundation for earbuds and small smart devices.
mpWWD detects the wake word while mpLocalization uses multi-microphone input to estimate the direction or position of the voice.
mpAB processes store noise and prompt echo, mpASR converts speech to text, and mpLLM interprets menu, options, quantity, and context.
The real ordering API and target hardware still need validation.
mpS and mpDiarization process multi-party input and separate per-speaker segments.
Connected to mpASR, this extends to per-speaker transcripts.
Yes.
mpNC, mpAEC, mpBeamforming, and mpAB apply in front of your existing ASR, improving the input audio.
No.
Depending on microphone count, speakers, user position, number of talkers, and required features, you can select only what you need.
Yes.
Starting from the usage scenario and required voice features, we can review microphone structure, the technology combination, and data collection conditions.
No.
Robots, kiosks, mobility, factories, and hearing assistance have deployment or validation experience in official material. Smart devices, meetings and voice chat, and defense are presented as capabilities and expansion directions.
Design Your Application Stack
Tell us your product type, microphone count, speaker structure, dominant noise, and the features you want to build — we will review the right mpWAV stack.
From single-mic noise control to wake·location·speaker analysis, multi-channel preprocessing, on-device ASR, and a conversational LLM — connected to fit your product environment.
Product Environment → Required Voice Technology → PoC → Platform Integration → Real-World Application
Robot · Kiosk · Mobility · Earbuds · Meeting · Factory
mpWAV develops and licenses voice interface technology that makes speech clear in noisy, real-world environments, built on 25+ years of speech signal processing research.
(주)엠피웨이브·사업자등록번호 859-81-01905·대표 박형민
Teilhard Hall 405, 35 Baekbeom-ro, Mapo-gu, Seoul, Korea
© 2026 (주)엠피웨이브. All Rights Reserved.
