Centre for Speech Technology Research (CSTR) is the website of an interdisciplinary research centre at the University of Edinburgh engaged in the study and application of speech recognition and synthesis technologies. The Centre publishes research papers and develops software for academic and commercial users. It also offers postgraduate research facilities. The CSTR website describes the Centre's objectives and the projects they are involved with. These projects are grouped into three principal areas: speech synthesis; automatic speech recognition; and database collection and labelling. Separate pages are devoted to each project within these fields. There are also sections for miscellaneous projects that do not fit into these main areas, including: named entity extraction; acoustic to articulatory inversion using neural networks; and voice transformation (or voice morphing). Several pieces of software may be downloaded from the site. These include the Festival Speech Synthesis System and the Unisyn lexicon. Research papers published by members of the CSTR may be downloaded from the site in PDF or Postscript formats. There is also news of upcoming workshops and conferences.
The Edinburgh University Speech Timing Archive and Corpus of English (EUSTACE) is a speech corpus comprising 4608 spoken sentences recorded for speech timing research at the department of Theoretical and Applied Linguistics at the University of Edinburgh. The full corpus is available for downloading and is intended to be useful for phonetics researchers and speech technologists working on synthesis and recognition. Example sentences are available for playback on the website, together with documentation including details of the experimental design, recording procedure, labelling methodology and original research results. The complete archive, available for downloading, includes a structured list of the sentences, the speech recordings and the label files, plus full documentation. Speech waveform files are available in WAW (RIFF) format and SD (ESPS) format. The downloadable corpus is free, and licensed for non-commercial use only. The original research was funded by the Engineering and Physical Sciences Research Council (EPSRC) and the production of the website was funded by the Moray Endowment Fund of Edinburgh University.
TalkBank is an interdisciplinary research project based at the Carnegie Mellon University and the University of Pennsylvania and funded by a five-year grant from the National Science Foundation (NSF). TalkBank aims to foster fundamental research in the study of human and animal communication. Its efforts are currently focused on six disciplinary groups of research: Animal Communication; Classroom Discourse; Linguistic Exploration; Gesture and Sign; Text and Discourse; Technical Development. TalkBank provides primary materials (databanks, corpora) for linguistic analysis, as well as standards and tools for creating, searching, and publishing primary materials. Membership in TalkBank and abiding by its rules for the use of the resources is essential. The use of TalkBank is governed by the Gnu Public License (GPL).
The Web-SLS: European Student Journal of Language and Speech website contains a number of articles concerned with speech technology. The articles are written by students but have been through a reviewing process before being published. The texts are freely downloadable as PDF-files. The project is supported by International Speech Communication Association (ESCA); European Chapter of the Association for Computational Linguistics (EACL); and European Network in Language and Speech (ELSNET). Although the site claims to have been updated in April 2009 no article is younger than March 2002, furthermore, the links to supporting associations are out of date. Despite this, it is a valuable resource for anyone studying computational linguistics and in particular speech technology.