Skip to content



Current Grants


Prosodic Phrase in Current Spoken Czech: Meaning, Balance, Stochastic Patterns

The project will provide multifaceted description of prosodic phrasing in spoken Czech with regard to news reading, artistic narratives and spontaneous dialogues. The acoustic domain will comprise dimensions of time, amplitude, fundamental frequency and spectrum with the aim to translate the ascertained cues and patterns into phonetic/phonological categories. Further phonetic description will consider material balancing or symmetry derived mainly from syllable counts, numbers and distributions of lexical stresses and actual accents. Crucially, the linguistic domain will encompass syntactic and semantic analyses of an extensive corpus of spoken texts aided by semi-automatic parsing. The weight of individual factors will be tested in perceptual monitoring experiments with latency measurements and in computational modelling.

Coordinator: Doc. Ing. Matoušek Jindřich, Ph.D.
Investor: GAČR 21-14758S
Years:2021 - 2023

Using a multimedia monolingual dictionary for modern teaching of Czech

The goal of the project is fundamental qualitative innovation of the on-line Dictionary of current Czech using audio and pictorial information and new functionalities aiming at internalization of its use by learners of Czech (even as a second language) as well as by the public and sensorially handicapped. It will consist in:
1. preparing audio for the headwords and 2 synthetic voices to be used in reading explanations and examples in modalities for: regular speakers, foreign learners, the visually handicapped and hearing-impaired
2. morphological and thematic classification of headwords
3. creating a classified usage notes database
4. pictorial material integration
5. creating a tool for learning material preparation for pupils and secondary school students

Coordinator: Doc. Ing. Matoušek Jindřich, Ph.D.
Investor: TAČR TL05000546
Years:2021 - 2023

Transformers of multiple modalities for more natural spoken dialog

The goal of the project is the research of more natural spoken dialog systems based on the Transformer framework. Since Transformers could be used in sequence-to-sequence scenarios, their use in natural language understanding and natural language generation is common. We would like to focus on the cases where the input or output of a neural network is speech. To convert speech into semantic representation or dialog intents we will be using the speech recognizer as a black-box but we plan to develop methods and approaches to process speech lattices in the general Transformer or recurrent neural networks. The inverse process of generating speech from intents will employ the pre-trained Transformer models for language generation and the recent DNN-based speech synthesis architectures. The dialog management will use the attention neural mechanisms to keep track of the dialog state and to generate consistent prompts in an informal or conversational style. The challenging task of speech synthesis using the given speech style will be backed by the recorded corpus of conversational speech.

Coordinator: Doc. Ing. Matoušek Jindřich, Ph.D.
Investor: GAČR GA22-27800S
Years:2022 - 2024