Speech @ FEEIT

The Speech Group at the FEEIT, CMUS has a long and rich tradition and expertise in speech technology, specifically speech synthesis and recognition, as well as analysis and processing.

Recent news

• Promotion of the Suze speech synthesis voice for assistive technologies.
• Localization of Cboard - an application for assistive communication AAC.
• The work of Speech @ FEEIT presented in the show Science Spectrum on MTV 1.
• We are pushing the envelope on naturalness in Macedonian Text-to-Speech synthesis.
• Govorko - a new tool for assistive speech communication for children with disabilities.
• State-of-the-art results in Large Vocabulary Continuous Automatic Speech Recognition in Macedonian .

Application

• Text-to-Speech Synthesis
• Augmentative and Alternative Communication
• Automatic Speech Recognition
• Speaker Identification
• Speech Rate Reduction

Research

• evoc-learn: Early Vocal Learning
• ProsoBeast: Prosody Annotation Tool
• Macedonian Phonology
• ProsoDeep: Prosody Modelling
• SP2: SCOPES Project on Speech Prosody

Get in touch

We are always on the look out for the next challenge!

If you want to increase your impact by making your work speech-enabled, or if you are solving a problem that involves speech or audio - we can help!!

Partners and Friends

• Association for Assistive Technologies “Open the Windows”
• Institute of Macedonian Language “Krste Misirkov”
• UNICEF
• Gordian Systems
• Fund for Innovations and Technology Development
• GIPSA-lab, Grenoble-INP, Grenoble, France
• ReTell ltd, Sunbury on Thames, UK
• Idiap Research Institute, Martigny, Switzerland