Centre for Speech Technology Research (CSTR)

Description

The Centre for Speech Technology Research (CSTR) is an interdisciplinary research centre linking Informatics and Linguistics and English Language.

Founded in 1984, CSTR is concerned with research in all areas of speech technology including speech recognition, speech synthesis, speech signal processing, information access, multimodal interfaces and dialogue systems. Members have many collaborations with the wider community of researchers in speech science, language, cognition and machine learning for which Edinburgh is renowned.

CSTR names machine learning among its core current research areas alongside speech recognition, speech synthesis and dialogue systems, and its groups build deep neural network architectures for the recognition and synthesis tasks. Researchers at the centre developed and released Merlin, an open source neural network speech synthesis system, and have published work on deep neural network acoustic models for speech recognition, including multitask learning of context-dependent targets and simplified very deep convolutional neural network architectures for robust recognition. A further strand develops deep architectures for statistical speech synthesis that incorporate models of how speech is produced and perceived. Machine learning and neural network methods are therefore a core, technical component of the centre's speech technology research rather than a peripheral interest.

Currently active

Yes

Offers funding

No, this infrastructure does not provide funding.

Founding year

1984

Contact details

On the map

Go to larger version of this map

Categorisation

Type

Tags

University affiliation(s)

University of Edinburgh
Edinburgh