Low resource NLP
Chimege Systems is a Mongolian startup company founded in 2019 to develop AI solutions for low resource languages. We believe in order to use AI applications widely, the first and foremost thing to do is develop natural language processing. Every application will use NLP for human-to-computer interactions. Our mission is to implement a variety of applications such as robots, smart appliances, and smart cars. We are happy to develop NLP for other low-resource languages as well as consult based on our experience in the field for the past 20 years.
Chimege writer™ - From speech to text
Automatically converting speech or audio files to text files. Speech recognition for Mongolian was a dream until now. Our proprietary technology achieved human-level speech recognition. The accuracy of our speech recognition reached an outstanding result of 96% and we continue working on improving every day. We found that If we manage to improve the recognition rate by one percent, user productivity will increase by at least 10%.
Chimege Reader™ - From text to speech
Automatically reading texts with natural sounding voices. Our TTS (Text to Speech) model owns the best quality of synthesized speech for the Mongolian language. We are working on optimizations such as decreasing resources and increasing emotions to achieve true human-level speech.
Natural Language Understanding
Natural language understanding (NLU) helps computers understand and interpret human language by breaking down the elemental pieces of speech. NLU enables human-computer interaction and it is unanimously considered one of the hardest problems in the field of Artificial Intelligence.
We have started developing Mongolian language understanding technology based on our researches and experiences and already achieved some positive results.
Our team consists of expert, scientists, and passionate developers which make us true R&D company. Meet our leadership team.
Leads the software engineering and production team. He has been working as a software architect for 15 years. Badral graduated as a Master of Computer Science at RWTH-Aachen, Germany.
Leads the artificial intelligence team. He is a remarkable computer scientist and worked on many challenging AI projects last 15 years. Tuguldur graduated from the Technical University of Berlin, Germany.
Leads the data preparation team. She is an expert in linguistics. Uchral completed her Ph.D. in Corpus Linguistics at the National University of Mongolia.