The Speech Recognition Group is part of the Machine Intelligence Lab at the University of Cambridge. Its primary area of research is large vocabulary speech transcription. Its research interests also include spoken dialogue systems, multimedia documen...
Czech Technical University research focusing on enhancement of speech in the running car environment, speech recognition and to creating of databases of natural language. Demo available for doing spectral subtraction on your own data.
Investigates a method used to help developers verify their dialog models. Multimodal implementation for VoiceXML is given. Free binaries and source code.
Dedicated to speech technology research, development, and deployment. Offers open source speech software, speech and language projects, publications, and resources.
Conducts research in the areas of speaker recognition (identification and verification), signature verification, speech recognition and handwriting recognition. Recognition engines and applications are currently being developed.