AI/IOT
Service Overview
With 20+ engineers, our AI team offers comprehensive consulting and development services. We specialize in Data Science, Deep Learning, and Machine Learning. Our expertise extends to BPO data labeling for enhanced accuracy and cost efficiency. Proficient in Java, Python, Lisp, Prolog, we excel in auto-training and re-training models, with a track record of building projects from the ground up.
AI Service
- Text RecognitionOur technology seamlessly identifies text, images, and sounds using pre-defined data. We prioritize efficiency by employing OCR technology only when other methods encounter difficulty in reading specific text types, ensuring swift and accurate text processing.
- Image ProccesingWe use live camera feeds and advanced image processing, extracting customer parameters like age and gender. Employing open-source libraries (OpenCV, VXL) and machine learning frameworks (TensorFlow, PyTorch), we enhance accuracy with deep learning neural networks, including GANs and DNNs.
- Natural Language ProcessingAt BAP, our technical research focuses on the dynamic interaction between computers and humans. We specialize in using chatbots to extract valuable data, adeptly analyzing user hobbies and opinions from diverse sources, including blogs and social media platforms.
- Voice RecognitionWe specialize in a cutting-edge technique for processing natural language through sounds. Our expertise lies in voice recognition and summarization, enabling computers to comprehend human commands and seamlessly transform text data into spoken voices.
- Traing-cost optimizationAt BAP, we ensure secure personal information handling and leverage AI for cost-effective scaling on Kubernetes systems.
Speaker recognition and speaker verification
Problem: Speaker recognition: is the process of automatically recognizing who is speaking by using the speaker-specific information included in speech waves to verify identities being claimed by people accessing systems; that is, it enables access control of various services by voice. Speaker verification (also...
Image processing in automation and robotics
Problem: RUBIK’S CUBE SOLVING Solving a Rubik’s cube is a challenge for people. It is required a lot of patient and practice to learn how to solve a Rubik’s cube. The question is, can we solve this problem using robot? – INPUT: The picture...
Wines Detection
Problem: Get the name of wines which was take photograph by users. In this problem, we need to recognize a large number of wines (up to 10000 labels) Solution: To solve above problem, we using deep learning for bottle detection and embedding, KNN search...
Pothole Detection
Problem: Identify potholes on road surfaces used for warnings or self-driving cars. INPUT: The picture of road surface. OUTPUT: Pothole location. Solution: For this problem, we used image processing and deep learning for location anh detect pothole....
Using NLP to discover hot topics on social network
Problem: The data on social platform are growing everyday. How can we use such data to discover “What is happening” on Facebook or Twitter everyday or even every hour? We can use the topic modeling for this problem: – INPUT: A dataset of Facebook’s posts of the...
Smart OCR
Challenges: Knowledge is humanity’s greatest asset. Books are invented to keep the knowledge is well-shared from generation to generation. However, in the digital transformation age, we need to transfer the data from books to digital databases. If you are looking for a method to...