http://www.voxforge.org/home/forums/message-boards/googlesoc http://www.voxforge.org/home/forums/message-boards/googlesoc/-speech-technology-in-computer-aided-language-learning Here is an interesting paper (it's a bit old - 1998...) that might be a good project idea: http://www.voxforge.org/home/forums/message-boards/googlesoc/-speech-technology-in-computer-aided-language-learning Fri, 09 May 2008 08:17:34 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/remote-server-based-continuous-speech-recogniser idea from Sam: ... [this] idea is only of any use once you have enough hours of http://www.voxforge.org/home/forums/message-boards/googlesoc/remote-server-based-continuous-speech-recogniser Thu, 08 May 2008 10:24:22 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/ways-to-make-entering-speech-more-fun From a post by Sam: ways to make entering speech more fun, so people would be encouraged to do it. A few ideas: To speak something out loud is a great aid to learning. Maybe we can find some resource of historical & scientific facts song lyrics (may be a bad idea because people would sing instead of speak.. But maybe that would be OK??) Movie http://www.voxforge.org/home/forums/message-boards/googlesoc/ways-to-make-entering-speech-more-fun Thu, 08 May 2008 10:20:18 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/google-summer-of-code-2008---post-mortem Came across this page on the Google site today that would be helpful for next year's application: Notes on Organizations Selection Criteria Organizations for Google Summer of CodeTM are chosen each year from the http://www.voxforge.org/home/forums/message-boards/googlesoc/google-summer-of-code-2008---post-mortem Mon, 17 Mar 2008 19:46:00 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/how-to-get-more-voice-samples I don't know if this a viable solution, but might worth thinking about it a bit. The http://www.voxforge.org/home/forums/message-boards/googlesoc/how-to-get-more-voice-samples Mon, 17 Mar 2008 16:51:43 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/use-voip-to-collect-speech dano suggested the use of the Ekiga SIP client for collecting/contributing speech. Would need to set up Asterisk VoIP server on the VoxForge repository hosted server, using ports greater than 1024 (so we would not need root access to the server). The VoxForge IVR application might serve as a start to this application, but something similar to Voice2Type's approach might work too. See this post for additional information. Ken http://www.voxforge.org/home/forums/message-boards/googlesoc/use-voip-to-collect-speech Fri, 14 Mar 2008 20:02:23 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/communication-aid People with a hearing impairment have a hard time taking part in conversations with other people with good ears. As a rule most people are not abble to use sign language. Also people with severe hearing problems often aren't good at using proper intonation due to a lack of significant feedback. This results in strongly diminished potential to take part in e.g. round table discussions. In theory speech technology can solve these issues. The perfect program would offer the following: the use of speech recognition to convert the speech of the discussion partners into text; voice recognition to sort the text per speaker in group discussions and text to speech to reply using a keyboard or any other text input device The speech recognition part should be aware of the speech synthesis part. It would not be realistic to think that this can be accomplished by one student in one smmer, but it might be interesting to create a small part of it. Does open source voice recognition even exist? Anyway enough challenges in just one 'little' application. Robin http://www.voxforge.org/home/forums/message-boards/googlesoc/communication-aid Thu, 13 Mar 2008 13:20:00 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/applet-to-allow-users-to-validate-submitted-speech Users should be able to view a submisson on-line, and listen to audio to ensure that it matches the prompts, make corrections if necessary (to prompts, or minor audio edits), rate the submission (numeric rating, or make comments about quality ), and save their changes to the Subversion repository (for version management). http://www.voxforge.org/home/forums/message-boards/googlesoc/applet-to-allow-users-to-validate-submitted-speech Sun, 09 Mar 2008 21:23:39 -0500 http://www.voxforge.org/home/forums/message-boards/googlesoc/web-front-end-to-our-pronunciation-resources Pronunciation resources will never be complete, as new words and expressions emerge. This project should allow users of ASR and TTS software to individually propose solutions to their pet-problems in pronunciation and have it automagically be fixed the next day. A solution would implement a front-end (java applet or other) to allow for the manual addition of entries to a pronunciation resource from a queue of words with unknown pronunciation or for a word given by the user. The user could be helped in determining the right pronunciation by showing the relevant parts of a corresponding Wiktionary-entry, or the proposition of an automatic G2P (such as espeak's). Also, the user should have easy and intuitive means of entering the phoneme symbols. Ideally, pronunciations for all forms of a word would be registered at the same time, which means, that parts of speech have to be determined (either manually or from context using a POS tagger) and corresponding entry masks be shown. http://www.voxforge.org/home/forums/message-boards/googlesoc/web-front-end-to-our-pronunciation-resources Tue, 04 Mar 2008 17:02:48 -0600 http://www.voxforge.org/home/forums/message-boards/googlesoc/morphologically-rich-pronunciation-dictionaries Current pronunciation dictionaries are simple lists of words and their pronunciations. While this works well for morphologically simple languages such as English, this is unnatural, inefficient and unmaintainable for morphologically richer languages, such as all Roman languages, German, Slavic languages and even more so for agglutinative languages such as Finno-Ugric languages and Turkish. The purpose of this project is to extend the W3C Pronunciation Lexicon Specification allowing for different parts of speech, define their default derivations and to also support irregular forms if these occur for specific word types. Solutions should contain an application to define allowed parts of speech and their behaviour for different languages and an ideally flexible application that can be used to build pronunciation resources for these languages. http://www.voxforge.org/home/forums/message-boards/googlesoc/morphologically-rich-pronunciation-dictionaries Tue, 04 Mar 2008 16:42:41 -0600 http://www.voxforge.org/home/forums/message-boards/googlesoc/scripts-to-train-acoustic-models-using-audiobooks-from-librivox http://www.voxforge.org/home/forums/message-boards/googlesoc/scripts-to-train-acoustic-models-using-audiobooks-from-librivox Tue, 04 Mar 2008 00:14:46 -0600 http://www.voxforge.org/home/forums/message-boards/googlesoc/modify-the-speechsubmission-java-applet-to-use-flac We http://www.voxforge.org/home/forums/message-boards/googlesoc/modify-the-speechsubmission-java-applet-to-use-flac Tue, 04 Mar 2008 00:09:04 -0600 http://www.voxforge.org/home/forums/message-boards/googlesoc/dialog-manager---lite Create a Command and Control http://www.voxforge.org/home/forums/message-boards/googlesoc/dialog-manager---lite Tue, 04 Mar 2008 21:45:05 -0600 http://www.voxforge.org/home/forums/message-boards/googlesoc/using-games-as-a-way-to-get-users-to-submit-speech From Brough Turner's blog: There are three ways to create a collective work: 1. Pay http://www.voxforge.org/home/forums/message-boards/googlesoc/using-games-as-a-way-to-get-users-to-submit-speech Mon, 03 Mar 2008 23:56:43 -0600 http://www.voxforge.org/home/forums/message-boards/google-summer-of-code/language-models-for-dictation-applications Compile a Written Corpus that can be used for the creation http://www.voxforge.org/home/forums/message-boards/google-summer-of-code/language-models-for-dictation-applications Tue, 06 Mar 2007 13:33:51 -0600 http://www.voxforge.org/home/forums/message-boards/google-summer-of-code/acoustic-model-conversion-scripts-for-isip/htk/julius/sphinx http://www.voxforge.org/home/forums/message-boards/google-summer-of-code/acoustic-model-conversion-scripts-for-isip/htk/julius/sphinx Tue, 06 Mar 2007 13:33:07 -0600 http://www.voxforge.org/home/forums/message-boards/googlesoc/2007-software-client-to-permit-users-to-record-and-submit-their-speech-audio-to-the-voxforge-web-site VoxForge needs the ability to let users record their speech and submit the audio (in uncompressed wav format) to http://www.voxforge.org/home/forums/message-boards/googlesoc/2007-software-client-to-permit-users-to-record-and-submit-their-speech-audio-to-the-voxforge-web-site Mon, 25 Feb 2008 13:58:00 -0600