Child Language Development AI
Our voice data-based child early diagnosis AI solution utilizes speech recognition, natural language processing, and machine learning technologies to detect language development delays early and provide customized support. As an efficient diagnostic tool for parents, teachers, and speech therapists, it's a solution for all children who need language development support and management.
Simultaneous analysis of voice and lip shape data for multimodal AI-based precise diagnosis (35% improvement in diagnosis accuracy)
Complete screening in about 5-10 minutes without expert checklist
45% improvement in diagnosis efficiency through acoustic analysis and sentence/vocabulary pattern automatic analysis in expert dashboard
Cloud-based automatic report generation and API integration with EMR and education office systems
Custom app development planned for parents and teachers with home-based intervention guides, development tracking, and expert connections
Subscription service model for continuous development management and customized content
Service Interface
* The actual interface is under development and the above image is an example.
Problems We Solve
1 in 10 infants experience language development delay, difficult to detect early due to shortage of experts
Existing diagnostic methods are time-consuming, expensive, and lack objectivity
Speech therapists manage 10+ children on average, making it difficult to maintain treatment quality
Lack of evidence-based support methods for parents leads to ineffective home-based intervention
Market Status and Outlook
Among 3.7 million children in Korea, about 370,000 are potential users of speech therapy services. Target customers include 12,000 speech therapists, 5,000+ speech therapy centers, and 30,000 daycare centers. The global speech therapy market is expected to expand with an average annual growth rate of 8.5% until 2030.
Core Technology
Large-scale Korean children's voice data collection and processing pipeline
Child voice specialized STT(Speech-to-Text) conversion model
Deep learning-based analysis algorithm for language development evaluation
Custom development pattern analysis and prediction engine
AI model trained on speech therapists' diagnostic expertise
Speech Therapist Dashboard
Efficient Treatment Management
View patient diagnosis results and development trends at a glance, and easily establish customized treatment plans.
Differentiation Points
Moving beyond traditional questionnaire-based evaluation methods, we provide integrated diagnosis of language areas including pronunciation, vocabulary, grammar, and pragmatics by analyzing children's actual voices. Based on AI algorithms trained on expert judgment and big data analysis technology, we provide highly accurate diagnostic results.
Development Roadmap
2023-2024
Starting February 2025
May-August 2025
November 2025
1. Data Collection and Analysis
2. Core AI Engine Development
3. Service Interface Development
4. Beta Service Launch
Service Launch:
Currently in the data collection and analysis phase, core AI engine development will begin in February 2025. We will present the optimal solution in November 2025 through collaboration with speech therapy field experts.
Art Information AI
Song & Stark's Art Information AI is an art information platform that objectively evaluates artist and artwork value and analyzes art market trends by collecting and analyzing artwork data using artificial intelligence technology. Through computer vision and deep learning technology, it comprehensively analyzes visual characteristics of works, artist styles, and market value to provide reliable information to collectors, gallerists, and art investors.
Artist style and artwork authentication support through visual characteristic analysis
Art market trend analysis through domestic and international auction/exhibition data analysis
Artist value evaluation based on career, exhibition history, and artwork transaction records
Artwork recommendation and gallery matching system through collector preference analysis
Art investment portfolio composition and value appreciation prediction analysis
Customized marketing insights for galleries and artists
Art Information AI Interface
* The actual service is under development and the above image is a beta version interface.
Problems We Solve
Low trust due to information imbalance and opacity in the art market
Lack of objectivity and consistency in artist/artwork value evaluation
High barriers to entry for new artists and market entry
Lack of objective indicators and analysis tools for art investment
Absence of efficient matching system between galleries and collectors
Market Status and Outlook
The domestic art market size is about 1 trillion won as of 2023, growing more than 10% annually, and the global art tech market is expected to expand with an average annual growth rate of 15% until 2025. Target customers include 700+ domestic galleries, 10+ auction houses, and 50,000+ art collectors. Demand for data-based services is rapidly increasing due to expansion of online art market and growing interest in art investment among MZ generation.
Core Technology
Computer vision-based artwork style analysis and classification system
Large-scale art database construction and management system
Machine learning model for market price prediction
Artist career analysis and evaluation algorithm
User preference analysis and recommendation engine
Gallery-collector matching algorithm
Artwork/Artist Analysis Dashboard
Comprehensive Art Information Analysis
Provides an intuitive dashboard by comprehensively analyzing artist's market position, artwork price fluctuation trends, style characteristics, and collector responses.
Differentiation Points
Moving beyond traditional qualitative and subjective evaluation methods, we provide quantitative and objective analysis using vast data and AI technology. In particular, we build specialized databases for Korean and Asian artists and emerging African artists to provide differentiated insights in the global art market. We provide more realistic and practical information by utilizing real-time market data connected to gallery operations.
Development Roadmap
January-June 2024
July-December 2024
January-September 2025
October 2025
1. Artocean.net Alpha Version Launch
2. Analysis Model Development
3. Platform Enhancement
4. Integrated Service Launch
Service Launch Information:
Currently in beta service through artocean.net, we will expand data scope and enhance algorithms until the first half of 2025 to officially launch the integrated service in October.
Song & Stark AI Technology
Advanced Language Processing
High-precision analysis system utilizing natural language processing and speech recognition technology
Data-Driven Decision Making
Accurate prediction and recommendation algorithms through big data analysis
Scalable Platform
Flexible and scalable AI platform applicable to various industries
