Existing ANC systems operate based on algorithms.

As a result, they are effective for patterned noise but have limitations in handling complex or highly variable noise.

To overcome these algorithmic limitations, we propose a new framework utilizing generative models.

This approach aims to mitigate the occlusion effect and enhance ANC performance in complex noise environments.

Designing & implementing AI model for BLE-based angle detection.

Employing Deep Learning approach (GRU & Linear Regression) by low level I/Q sample data.

Standardization : Submitting & having IPR of new specifications in Bluetooth SIG (DFWG)

Driver Emotion-Aware Personalized AI Music Composition

Generate Latent vector about melody, rhythm, chords, musical instruments

Build pattern-based classification model and relatedness-based clustering model

AI composition music based reinforcement learning

Generated music can be played longer, perform Reinforcement learning 

Using Double DQN or Dueling DQN

Build Music Relationship music Regressor for arrangement recommend and Emotion-based search

Define the MRF Score, that indicates the relationship between music

Classification about lightness, saturation, and color mean user's emotion

Define loss function about pitch and density in music

After pre-learning latent vector z in Music VAE, training z using Actor-Critic method with loss fuction

Music Source Separation & Automatic Music Transcription using Deep Learning

Fashion Component Classification using Mask R-CNN 

& Fashion Image Generating using GAN

Indoor position recognization under BLE RSSI enviroment using Deep Learning

(CNN, Auto-Encoder, ...)