AI + Education

Our convenient AI technology allows ESL teachers to focus on quality video creation instead of the nitty-gritty manual setup.

Speech recognition

Our company achieved a state-of-the-art speech recognition system performance using end-to-end deep learning. Through a combination of data augmentation, grammar correction, and text alignment technology, users can benefit from an effective in-app speech to text recognition and translation feature that fosters quality learning.

Scene Recognition

By bringing together our advanced scene recognition technology, tagging engine, and user-generated content (UGC, such as video, voice and text), BlaBla surpassed the industry network's performance on video scene interpretation (such as that of ImageNet and AlexNet) by 10 percent. The scene recognition technology is essential in the development of our user interest maps which assign tags to videos and categorize videos into various themes automatically.

User Interest Profile

We developed a strong interest profile with user interest data and natural language processing (NLP) by building a mining engine that matches users to videos that are suitable for their current English proficiency within the topics they selected. This provides a solid foundation for our recommendation system.

Recommendation Sytem

Unlike the industry's conventional recommendation system, BlaBla's recommendation engine combines multi-dimensional data such as audio, video, text, user interest maps, and real-time interest analysis based on user flow. Our advanced recommendation system provides accurate and optimized real-time interest matching and content recommendation outcomes.