LINGUIST List 8.563

Mon Apr 21 1997

FYI: Bell Labs Addition

Editor for this issue: T. Daniel Seely <seelylinguistlist.org>

Directory

  • Richard Sproat, New Addition to the Bell Labs Text-to-Speech Page

    Message 1: New Addition to the Bell Labs Text-to-Speech Page

     Date: Mon, 21 Apr 1997 14:45:36 -0400 (EDT)
    From: Richard Sproat <rwsresearch.bell-labs.com>
    Subject: New Addition to the Bell Labs Text-to-Speech Page


    We are pleased to announce a new addition to our text-to-speech (TTS) website, namely a page containing demos of TTS work done by summer students at Bell Labs.

    The address is:

    http://www.bell-labs.com/project/tts/student.html

    (The address for the main TTS page is http://www.bell-labs.com/project/tts/)

    The page contains some samples of the Navajo system done by Bob Whitman during the summer of 1993; and an interactive demo of the Romanian TTS system done by Karen Livescu during the summer of 1995.

    We expect to be able to continue to host students interested in working on such projects in the summers to come. We would be particularly interested in hosting students who are native, or near-native speakers of languages for which we have not previously done TTS work (i.e., most of the world's languages); and who have a background in linguistics, computational linguistics or relevant areas of engineering. Relevant computational experience --- some ability to program, and familiarity with Unix or Linux --- is also desirable.

    Text-to-speech synthesis is a technology that presents a rich array of challenges, and it requires one to understand a variety of areas including acoustic phonetics, prosody, computational phonology, computational morphology, and corpus-based linguistics --- to name only a few. Working on a TTS system is thus a very good way to get practical experience in several areas of computational speech and language modeling. Anyone with the above-mentioned qualifications who is interested in spending a summer working with us should feel free to contact any member of our group. (Note that at this point it is too late for 1997: a good time to think of applying is anywhere from November through January for the following summer.)

    For the Bell Labs Text-to-Speech group:

    Richard Sproat

    - Richard Sproat Language Modeling Research Department Multimedia Communications Research Laboratory Bell Laboratories, Lucent Technologies | tel (908) 582-5296 700 Mountain Avenue, Room 2d-451 | fax (908) 582-3306 Murray Hill, NJ 07974, USA | rwsbell-labs.com http://www.bell-labs.com/project/tts/