Artificial Intelligence in Speech Modelling
In Russian
Level of education Master
Type of instruction Full-time
Duration 2 years
The programme is aimed at training specialists in artificial intelligence and big data in speech technologies. Graduates of the programme will be specialists in a wide range of areas: methods and algorithms for processing large amounts of audio materials to automatically generate speech messages; extraction of content information from a voice signal, in particular information about the speaker (social characteristics, physiological, mental state, etc.), verification and identification of the speaker; designing of conversational agents and human-machine speech-based interface; linguistic aspects of artificial intelligence.
Description and modeling of language (speech and text)
- Formal Methods for Speech Modelling
- Phonology and Speech
- Psychological and Physiological Aspects of Speech
- Acoustic Analysis of Speech Signals
- Segmental and Suprasegmental Units of Language
- Phonetic Universals
- Types and Forms of Speech
Courses in Machine learning
- The Mathematics of Machine Learning
- Probability and Statistical Models
- Machine Learning
- Neural Networks and Deep Learning
- Big Data Analytics
Courses in language processing
- Automatic Speech Recognition
- Automatic Speech Synthesis
- Human Identification by Voice
- Analysis of multimodal data in identification of psycho-physiological state of individual
- Speech and Multimodal Data Corpora Design
- Automatic Transcription
Courses in software engineering
- Python Programming
- Neural Networks Programming
- Text Processing and Speech Technology
- Digital Processing of Speech Signals
- Programming of Text Generation and Processing
- The programme is designed to train highly qualified specialists who can: apply both the latest mathematical apparatus used to design of AI systems and knowledge in the field of linguistics, psychology and physiology; find solutions to current scientific and technical problems in business and academic research. Graduates of the programme will acquire both practical skills in AI problems solving and a theoretical knowledge of linguistics, which will allow them to further improve their qualifications
- Among the lecturers of the programme are specialists who lead and participate in research projects related to speech modelling and design of speech technologies commissioned by major government funds and companies. Lecturers of the programme are acknowledged experts in this field. They are members of the International Speech Communication Association, programme committees of key international conferences on speech technologies and linguistics (Interspeech, ICASSP, LREC, Speech Prosody, ICPhS, Specom, etc.)
- Automatic Speech Recognition
- Automatic Speech Synthesis
- Human Identification by Voice
- Analysis of Multimodal Data in Identification of Psycho-physiological State of Individual
- Speech and Multimodal Data Corpora
- Design Design of Speech Technologies for Under-Resourced Languages
- Design of Conversational Agents
- Indian Institute of Technology, India
- Ludwig Maximilian University of Munich, Germany
- Brno University of Technology, Czech Republic
- International Speech and Communication Association
- Linguist
- Teacher
- Researcher
- Programmer
- Statistitian
- Data analysis specialist
- Information Technologies (IT) Senior Project Manager