http://www.voxforge.org/home/forums/message-boards/general-discussion http://www.voxforge.org/home/forums/message-boards/general-discussion/new-iphone-voice-dialing-apps Here are a couple of new apps for the iPhone that use Open Source speech recognition engines. Although the apps themselves are closed source, it is interesting to see what can be done with PocketSphinx and Julius. http://www.voxforge.org/home/forums/message-boards/general-discussion/new-iphone-voice-dialing-apps Mon, 29 Sep 2008 21:56:17 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/legal-information Hi everyone, http://www.voxforge.org/home/forums/message-boards/general-discussion/legal-information Tue, 23 Sep 2008 13:47:45 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/problems-with-registration-and-password-reset I updated the VoxForge website to the most current version of the WebGUI (r7.5.22-stable) content management system this past Sunday (Sept 14). Unfortunately, this broke the registration and password reset. http://www.voxforge.org/home/forums/message-boards/general-discussion/problems-with-registration-and-password-reset Wed, 17 Sep 2008 10:42:46 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/translate-the-voxforge-applet-to-your-own-language You can translate the speech submission applet of VoxForge and make the first prompts for your language now at http://translations.launchpad.net/voxforge Daniël http://www.voxforge.org/home/forums/message-boards/general-discussion/translate-the-voxforge-applet-to-your-own-language Tue, 09 Sep 2008 09:16:12 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/freeclas---free-commons-of-linguistically-annotated-speech From the a comp.speech.research post: FreeCLAS (http://www.ihear.com/FreeCLAS) is a new project to build a a data base of high-quality speech data. "High quality" means annotated data that have been validated by humans. Building such a data base has been expensive because it requires substantial investment of people's attention. As a result, high-quality speech data is not generally available. FreeCLAS uses a wiki. This is a call for people to join the wiki to build it. Embedded in the wiki is a tool, shva, which opens from your browser to let you hear, view and annotate any utterance in FreeCLAS. At this point, there is a seed data base of a small collection of utterances annotated in en-US and IPA. shva and other related software downloadable from FreeCLAS are all Free Software, licensed under GPL or other compatible licenses. The speech data is under the Creative Commons attribute-share-alike license. Their focus seems to be more collecting linguistic annotations of speech by getting users to provide/validate time stamps of utterances. This is a little different what VoxForge is doing. We are basically trying to collect speech prompts (15-20 words long), with little regard for accurate timings - since the HTK/SPhinx acoustic model training process can do this automatically (with short utterances) What is really interesting (from VoxForge standpoint at least) is their ALingA (GPLv3) annotation Java applet. I can't get the app the run on my PC (I have a 64-bit machine, which they don't provide support for...yet). However, from the screen shots, it looks very impressive for a Java applet. They use the JavaFX libraries, which is Sun's answer to creating rich Internet applications (RIAs)... i.e. Sun approach to creating a Flash-like environment. It might be a useful starting point for a speech submission annotation validator for VoxForge (but just to allow other users to validate that an utterance matches the prompt line). Ken http://www.voxforge.org/home/forums/message-boards/general-discussion/freeclas---free-commons-of-linguistically-annotated-speech Thu, 04 Sep 2008 12:33:52 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/making-voxforge-corpus-useful-for-asr-research Hello, I am posting to share some thoughts regarding ASR research and the planned 1.0 release of the VoxForge corpus. The goal of VoxForge is to create speech corpora for use by the FOSS http://www.voxforge.org/home/forums/message-boards/general-discussion/making-voxforge-corpus-useful-for-asr-research Wed, 03 Sep 2008 21:42:55 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/acoustic-model-0_1_2 When do we release 0.1.2? Ticket 202 (20% of 140 goal is fixed, according to the metrics page we have now 36%, but misses a month of speech submissions => 37%, 38%.) http://www.dev.voxforge.org/projects/Main/ticket/202 I thought 376 (http://www.dev.voxforge.org/projects/Main/ticket/376) is fixed (according to the forum thread.) I don't know about other tickets but 366 http://www.dev.voxforge.org/projects/Main/ticket/366 doesn't seem to be a showstopper for the English acoustic model. "Update Acoustic Model creation scripts and Tutorials (and Howtos) to Julius 4.0" which is supposed to acoustic model 0.1.3, I think that has a bigger priority to my opinion. Let me know what you think about it. (What was it again: "release soon, release often" isn't it? http://www.voxforge.org/home/forums/message-boards/general-discussion/acoustic-model-0_1_2 Mon, 01 Sep 2008 14:50:43 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/different-depths-of-voice Hi there, http://www.voxforge.org/home/forums/message-boards/general-discussion/different-depths-of-voice Wed, 27 Aug 2008 21:34:09 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/processing-of-speech-submissions-delayed2 I am travelling for the first 3 weeks of August. Because of this, processing of audio submissions will be delayed until I return (if I have time, I may be able to do some of these remotely). You can still submitt speech - it will just collect on the submission server. I will process all submissions when I get back. thanks, Ken http://www.voxforge.org/home/forums/message-boards/general-discussion/processing-of-speech-submissions-delayed2 Sat, 09 Aug 2008 23:23:07 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/page-after-submitting Hi Ken, http://read.voxforge1.org/r0_1_4/endpage.php is currently in English and links again to http://www.voxforge.org/home/read. I've created the Dutch translation. Bedankt voor je bijdrage! Hieronder is de lijst te zien met bijdragen die momenteel wachten om opgenomen te worden in het corpus van VoxForge. Je spraak zal worden bewerkt en gecontroleerd worden om in het proces van deze avond worden opgenomen. Om de tijd te zien die je aan spraak hebt bijgedragen en hoe dicht we bij ons doel zijn kijk op [url=http://www.voxforge.org/home/downloads/metrics]de statistieken van de spraak-bijdrage[/url] (alleen nog in het Engels.) VoxForge's streven is om minstens 10 - 15 minuten spraak per spraakdonor te verkrijgen om aan ons doel van 140 uur te voldoen voor de eerste versie van de spraakcorpus en akoestische modellen van VoxForge. Om een goede dekking te leveren van de taal hebben we honderden verschillende zinnen aangemaakt. Maak je geen zorgen als je een aantal zinnen al hebt gedoneerd, dat is ook erg handig, een persoon zegt immers nooit twee dezelfde dingen op precies dezelfde manier! [url=http://voxforge.org/nl/read]Klik hier om opnieuw een bijdrage te leveren![/url] En vertel alsjeblieft ook vrienden en familie over VoxForge en vraag ze om donor te worden! Bedankt namens het VoxForge-team. -- http://www.voxforge.org/home/forums/message-boards/general-discussion/page-after-submitting Sat, 09 Aug 2008 07:54:51 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/developing-grammar Hi, I work in a project for create an english course online. we are using julius + voxforge for reading exercise . Basically, I created a new file .voca and .grammar, then i compile with mkdfa.pl for generate the files .dfa .dict .term. So, i used these files for recognise the phrase that the student pronounce, then i compare the phrase spoken with the phrase tha i was waiting. The english exercise is an phrase that student listening and after he pronounce the phrase and the julius + voxforge recognise the phrase, if the phrase recognized is equals the phrase listening the exercise is correct. Then, i would know if anybody created grammar, because my grammar is very large and i want tips for otimaze it. Thanks http://www.voxforge.org/home/forums/message-boards/general-discussion/developing-grammar Wed, 06 Aug 2008 12:25:54 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/writing-a-command-and-control-application-with-voice-recognition The Bloc d’en RainCT blog has a post describing a simple program to control Rhythmbox with your voice on Ubuntu. It uses the Julius speech recognition engine and a VoxForge acoustic model. http://www.voxforge.org/home/forums/message-boards/general-discussion/writing-a-command-and-control-application-with-voice-recognition Sun, 27 Jul 2008 08:51:54 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/application-java Hi, I am developing a application in java for speech recognition and i'm using voxforge with julius for this. Anybody here already do this? thanks... http://www.voxforge.org/home/forums/message-boards/general-discussion/application-java Tue, 22 Jul 2008 08:41:36 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge-user-submissions Is it time to grow the volunteer base for voxforge? From what I've read about the problem, what you need is lots of people to record their voice so you can produce acoustic models. Now if you really need people why don't you try and team up with the WikiMedia Foundation, if you can get even a small amount of wiki people to help out you'd vastly increase your incoming audio. That is of course if you do need more people. You may just be looking for more programmers with better ideas for the software and processing side of things. http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge-user-submissions Mon, 14 Jul 2008 22:11:03 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/untitled btw i run Firefox (but I tried it also with IE on Win XP) http://www.voxforge.org/home/forums/message-boards/general-discussion/untitled Mon, 14 Jul 2008 13:45:48 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge-submission-applet-does-not-work-due-to-router With some routers you can not upload recorded data via the Submission Applet because of the default firewall configuration in the router. I've tested it (I've a Thomson router). Is there a way to fix that in the Submission Applet? http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge-submission-applet-does-not-work-due-to-router Thu, 10 Jul 2008 12:00:15 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/myvox-telephony-to-voip-gateway I just came across the MyVox website, which "turns any phone into a microphone hooked up to your application". It's ad supported, and the audio ads are a little "over-the-top" (although they are short: 5-7 seconds), but it is a very interesting model that might finally get the "Voice Web" up and running with things like speech recognition based web searching, etc., without the need to invest serious money in telephony infrastructure. This might also be interesting from a VoxForge perspective as another alternative to collect speech (maybe modifying trevarthen's VoxForgeIVR app to perform this task). Ken http://www.voxforge.org/home/forums/message-boards/general-discussion/myvox-telephony-to-voip-gateway Thu, 19 Jun 2008 11:29:26 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-do-you-control-license-violation I would like to get your input on the following. As far as I understand, the GPL license doesn't allow non-free derivative works out of a work licensed under GPL. However, I believe that it is impossible to know whether an acoustic model has been compiled out of voxforge's audio corpora. Basically, the creation of an acoustic model requires: pre-processing -> feature vector extraction -> classification For example, an Hidden Markov model is composed of state transition probabilities and of pairs of means and variances for the observation probability distributions... There's no way to be sure that an acoustic model comes from voxforge's audio corpora and the commercial product in question will never have to ship the audio corpora, only the acoustic models... In this respect, how do you control that a commercial product doesn't use your corpora? Thanks for your input Mathieu http://www.voxforge.org/home/forums/message-boards/general-discussion/how-do-you-control-license-violation Wed, 28 May 2008 09:25:56 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/blizzard-2008-listening-tests-are-open In order to better understand and compare research techniques in building corpus-based speech synthesizers on the same data, the Blizzard Challenge has been devised. The basic challenge is to take the released speech database, build a synthetic voice from the data and synthesize a prescribed set of test sentences. The sentences from each synthesizer will then be evaluated through listening tests. We're pleased to announce that the Blizzard 2008 listening test is now available. We need your help in getting as many subjects as possible to participate. There are different start pages for various listener types - please take care to use the correct one: English ====== Speech experts: http://groups.inf.ed.ac.uk/blizzard/blizza...egister-ES.html Volunteers: http://groups.inf.ed.ac.uk/blizzard/blizza...egister-ER.html Mandarin ======= Speech experts: http://groups.inf.ed.ac.uk/blizzard/blizza...egister-MS.html Volunteers: http://groups.inf.ed.ac.uk/blizzard/blizza...egister-MR.html Subjects do not need to be native speakers - we gather information about this in the questionnaire at the end of the listening test. The listening test can be completed in one session, or over multiple sessions. If you speak both English and Mandarin, then you may participate in both listening tests if you wish. Please publicise this as widely as possible on your mailing lists, blogs, web pages, or whatever. Also, please ensure that as many members of your research group, other colleagues, students, family, etc participate as possible! We need hundreds of subjects for each language. Remember to direct them to the appropriate start page. Note that we are also running several other listener groups for paid subjects; if you would like to organise a group of paid subjects, then please contact me. We can also set up groups for other specific listener types, if we know there is a large enough pool of available subjects with certain characteristics; contact me if you want to discuss this. Please report problems to blizzard@festvox.org http://www.voxforge.org/home/forums/message-boards/general-discussion/blizzard-2008-listening-tests-are-open Sun, 18 May 2008 12:01:25 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/mail-server-problems---fixed For the past few of days we've been having problems with the smtp server used by the VoxForge website. Email notifications of new posts to forums on the VoxForge site were not being sent. I've changed smtp relay service providers, so this problem should now be fixed. Ken http://www.voxforge.org/home/forums/message-boards/general-discussion/mail-server-problems---fixed Sat, 17 May 2008 13:39:25 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/querying-a-database-using-open-source-voice-control-software There is an article on Linux.com (written by Colin Beckingham) that outlines the steps the author took to create a small system that can query a database using speech ... with help from HTK, Julius, Audacity, Festival and the VoxForge tutorial and howto. From the introductory paragraph of the article: http://www.voxforge.org/home/forums/message-boards/general-discussion/querying-a-database-using-open-source-voice-control-software Fri, 16 May 2008 12:21:32 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/conversation-with-richard-stallman-re-voxforge-and-fsf From Richard Stallman: The Free Software Foundation and the GNU Project would like to help http://www.voxforge.org/home/forums/message-boards/general-discussion/conversation-with-richard-stallman-re-voxforge-and-fsf Wed, 14 May 2008 08:58:30 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/readying-voxforge-for-high-traffic Here is a transcript of an e-mail I have sent to the following just now: Ken McLean, maintainer, VoxForge project CC'd to: Richard Stallman, founder of GNU/Linux, FSF David Huggins Daines, current sphinx maintainer Nickolai (nshm), sphinx developer Hi Ken! My name is Sam. I'm working on redesigning HCI, both at work and in my free time. I'm concerned that GNU/Linux doesn't as yet have any support for continuous speech recognition. I think that within the next 10 years there will be a major shift towards speech recognition, and I think it is important that GNU/Linux is not left out. I am in touch with the sphinx development community, including David Huggins Daines, the current sphinx maintainer. David confirmed that the problem is that there is no sizeable speech database (lots of .wav phrases together with their associated .txt), and hence sphinx cannot generate a decent voice model. I am also in touch with Richard Stallman, founder of GNU/Linux and the FSF. He is willing to publicise VoxForge to the FSF community. This will hit a lot of people, and there are a lot of linux users who really want to contribute but don't know how to code. This could generate a lot of traffic for VoxForge. Can VoxForge handle it? We should discuss this before throwing the gates open. Here are three things that hit me straight away: Firstly, To get people to contribute, it is important to have some simple feedback system. It is the difference between one one hand laying a brick, which disappears, and being told that one day a castle will appear, and on the other laying a brick on a partially built castle. Is there any chance you could include a usage graph on the website main page? x-axis: each pixel is one day y-axis: the number of phrases contributed that day And a thermometer! You know like those thermometers they use at fundraisers? This is how much we need to make a continuous speech recogniser, this is how much we have got... The key is to create a 'we can make it happen' vibe... Once people see the thermometer is starting to heat up I'm sure there will be a lot of people who put hours of effort in. Secondly, I just tried it out, imagining I'm a linux fan who has just seen an article by Richard Stallman in a linux magazine. I log onto Voxforge.org ... I didn't get very far. A dialogue box appeared telling me 'the page you're viewing requires java. More information is available on the Microsoft website.' And that was it. The page it takes me to has a link saying 'Information on the Java Security Warning pop-up', and this is quite a long page with a lot of information. It doesn't offer any solution to my problem; I have not been presented with any option to download the java virtual machine and. So I know better than try to get anything meaningful out of the Microsoft website! I go to google, put in 'download java virtual machine vista'... And have to take it from there. But this is going to put off a lot of people, maybe >90%. Is it possible for the browser to ascertain whether java is installed or not, and if it isn't, offer a link straight to downloading the appropriate java virtual machine executable for the operating system that person is using? ie minimize the amount of clicks and reading required... The third issue is the phrases themselves: where do you get them from? Nickolai (a major sphinx contributor, cc'd) and I were discussing 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 scripts. I swear, if you find a good movie script (like Star Trek IV) you will get people who read through the entire movie. Making Voice-books. people can kill two birds with one stone - they can read in a document, or a chapter from a book, creating a Voice-book while adding to the database. Have two text boxes: URL[ ], starting from [ ] So if I put in URL[http://www.chordie.com/chord.pere/www.ultimate-guitar.com/print.php?what=tab&id=456256], starting from [I'm afraid] It starts presenting text from this location, one sentence at a time. hit spacebar to advance. Of course you may need several people to speak the same phrases. If this is true, these ideas could be adapted: you could have a pool of 'this is what the last hundred visitors chose to read out', and next to each one, a number which represents how many people have read from that source. So you can either click something existing, or choose something new. This could be a lot of fun - who knows what songs / movies / literature / jokes people are going to put up? Sam (sunfish7@gmail.com) http://www.voxforge.org/home/forums/message-boards/general-discussion/readying-voxforge-for-high-traffic Tue, 06 May 2008 11:40:04 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/text-to-speech2 Would using a text to speech engine to read documents be a legitimate source of generating audio instead of manually reading it? http://www.voxforge.org/home/forums/message-boards/general-discussion/text-to-speech2 Thu, 24 Apr 2008 17:09:31 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/donations Have you considered taking donations, and then using that to fund voice samples through amazon's Mechanical Turk system. I figure that a reasonable low wage of £3(sorry i'm in the uk, i don't know what a reasonable low wage is in the US) per hour of audio, would get interest. I guess it depends on whether there's enough interest for enough donations to be made to fund a significant job on mechanical turk. http://www.voxforge.org/home/forums/message-boards/general-discussion/donations Fri, 18 Apr 2008 14:21:00 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-app-engine I've been playing with the Google App Engine SDK since it has been released (and am on the waiting list to get on the actual site). It runs Python, and can use the Django framework. It has a 500 meg storage limitation. From the Google App Engine blog: Google App Engine -- a developer tool that enables you to run your http://www.voxforge.org/home/forums/message-boards/general-discussion/google-app-engine Tue, 15 Apr 2008 20:13:04 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/language-models-for-speech-and-ocr-grammar-checkers I have been looking at speech recognition and OCR lately. I also read that language models can help speech recognition engines determine the most likely result from ambiguous input. It occurred to me that this is similar for OCR - OCR guesses letters and then has to determine which word is most likely based on what it thinks it "saw" and what the word is most likely to be. Grammar checkers in word processors must also determine the likelihood of entered text. This may be an off-the-wall suggestion, but would it be sensible for FSF to try and develop a GPLv3 language model that could be used for all three? I would have thought that a good language model was something important to Voxforge's aims (the ability to create speech recognition apps without the need for commercial resources). http://www.voxforge.org/home/forums/message-boards/general-discussion/language-models-for-speech-and-ocr-grammar-checkers Fri, 28 Mar 2008 21:27:22 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/keep-voxforge-alive We need people to do this, to keep VoxForge alive: #1 Blog about VoxForge updates, link to this website on your site, make a screencast, etc etc #2 Tell your friends about VoxForge, tell them to submit some speech if they have some time (not only English people, we need speech from other languages too.) #3 Show your friends Gnome Voice control / Sphinx or something. #4 Submit speech by yourself, or develop things that are important for VoxForge, take a look on the GSoC ideas. #5 Make VoxForge popular by doing things you think they are good! http://www.voxforge.org/home/forums/message-boards/general-discussion/keep-voxforge-alive Tue, 25 Mar 2008 08:00:17 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/links-on-news I don't see it was discussed somewhere btw, it would be nice to have a recommended reading page/thread. It would be nice to have a few documents that can get newbie into speech technology quickly. For a start I suggest to use David Gelbart's collection: http://www.icsi.berkeley.edu/~gelbart/edu.html But the question is a bit different, what resources do you use to track recent news in speech technologies? Probably some feed or blog is a good place to read. Something like http://www.speechtechblog.com for example http://www.voxforge.org/home/forums/message-boards/general-discussion/links-on-news Sun, 23 Mar 2008 03:21:49 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/things-to-improve-voxforge Hi all, I am not a technical guy but I do see lot of short-comings for community participation & the lack of it. It might sound unpleasant or whatever but this is my take on stuff :- 1 . Forums need a usability shot :- The forums are in a bad bad shape. I don't know what forum software are u guys using, but its just not something which many people could use/understand. Most of the forum softwares have a 'search' query thing where people can know if somebody has asked something before, so things are more organised. It should be visible all the time. 2. FAQ :- This FAQ should be visible all the times. An FAQ which answers all the oft. repeated queries such as :- i. How can I record the sounds? ii. Which softwares do I need to record the sounds? iii. How many sound clips are needed? iv. from which countries? v. what accents? vi. any particular mike or hardware which would be useful? and so on & so forth. This would make for huge gains. 3. There needs to be a blog which is accessible from the top itself so people know what new improvements are happening. 4. Break the requirements into small doable targets which are in the form of graph. Also tell what improvements would one have when we touch that target. 5. Make frequent releases of the corpus done & interact & blog the resulting improvements made in the various open source speech recognition engines. 6. Lastly, give alternative ways to do ftp submissions to the site. Give some generic instructions for people using console-based or graphical ftp clients to upload stuff. Perhaps there could be a way to tag them so they are attached to the job number automatically. Feel free to suggest & improve the suggestions :) http://www.voxforge.org/home/forums/message-boards/general-discussion/things-to-improve-voxforge Sat, 22 Mar 2008 00:06:51 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-2008---status Unfortunately, we did not get accepted to this year's Google Summer of Code 2008 project (an amazing program that offers student developers stipends to write code http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-2008---status Mon, 17 Mar 2008 19:36:10 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/recognise-its-a-voice-without-needing-to-know-whats-said Hi all. Im looking for something that just needs to recognise when something is being said (not what is said). Not quite silence detection because I want it to be able to distinguish between music and talking. A bit of a tall order? Any views or input greatly appreciated Dylan http://www.voxforge.org/home/forums/message-boards/general-discussion/recognise-its-a-voice-without-needing-to-know-whats-said Sun, 16 Mar 2008 07:30:37 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/sentences-from-opentaal The project OpenTaal (at opentaal.org) has collected a huge amount of Dutch sentences. OpenTaal is a Dutch project for creating dictionaries, grammar checking, synonyms and much more is coming. They made also 'Wordsharvester', a little app that collects and counts words from all over the web. Here is the link (more than 300 MB): http://opentaal.org/opentaalbank/test/zinnen.tgz They have also made a collection with the most used combinations of words (2,3,4 and 5 words), but that's currently not accessible due to a mysql error. I would greatly thank OpenTaal for the work that they did (and do)! Well, my question is, what can we exactly do with this huge information? How can we implement the information in the best way? Are there any ways of doing this yet? http://www.voxforge.org/home/forums/message-boards/general-discussion/sentences-from-opentaal Sat, 15 Mar 2008 12:20:16 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/untitled3 A new thing. http://voxforge.org/uploads/Ve/ei/VeeirBUXntnzL2oUm4s59A/Tekening.svg Open it with inkscape, in ff2 it looks nasty, ff3 crashes http://www.voxforge.org/home/forums/message-boards/general-discussion/untitled3 Fri, 14 Mar 2008 03:38:17 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/untitled4 A new one http://voxforge.org/uploads/yr/1r/yr1rPgM2e9PT7OYSztZiuQ/shutupcomputer.png http://www.voxforge.org/home/forums/message-boards/general-discussion/untitled4 Fri, 14 Mar 2008 16:06:08 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/banner-/-slogan I made this one http://voxforge.org/uploads/_5/X7/_5X7KruGxG-NxJkduIQEEg/8015banner.png Maybe a slogan: "Voxforge, because you want it to listen!" Sounds nice by me but I am not a native English speaker :) http://www.voxforge.org/home/forums/message-boards/general-discussion/banner-/-slogan Thu, 13 Mar 2008 13:58:51 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/ekiga Feature request: use Ekiga for collecting/contributing speech. http://www.voxforge.org/home/forums/message-boards/general-discussion/ekiga Tue, 11 Mar 2008 11:14:17 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-2008---application http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-2008---application Mon, 03 Mar 2008 23:25:21 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge-buttons-and-banners-and-logos-and-slogans Is there already a page with buttons and banners for VoxForge Like http://www.spreadfirefox.com/?q=affiliates/homepage that people and websites can put it somewhere? The users of Firefox have also increased by this form of advertising and so can VoxForge. By the way, I prefer a more clear logo. Like a microfone http://www.midigraphics.co.kr/upload/img/product/Beta58.jpg or a speaker http://www.webgraffix.com/PSPImages/Speaker.jpg http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge-buttons-and-banners-and-logos-and-slogans Tue, 04 Mar 2008 10:21:38 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/training-speaker-dependent-models-from-scratch I have some english audio lectures from my professor at school I want to use Julius as a continuous dictation ASR to create some rough transcripts for. The lectures have technical terms that a general english language model/dictionary will not suffice. How should I start (from scratch)? What's existing in the public domain that I can use? How do I create a language model? How do I train an acoustic model? How much transcribed data do I need? How much training and testing data should I prepare? http://www.voxforge.org/home/forums/message-boards/general-discussion/training-speaker-dependent-models-from-scratch Sat, 01 Mar 2008 17:53:01 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/use-of-the-quickstart-nightly-package-with-a-htk_acousticmodel-nightly Hello there, I have been using the quickstart linux nightly package for a few days and have ran into problems trying to use the HTK_AcousticModel nightly. From what I can see, the nightly Julius quickstart package contains the hmmdefs and tiedlist from the HTK_AcousticModel already. But julius is still using the sample.dict and sample.dfa of just 23 words. I would like to use the larger 11,000 word dictionary from the HTK_AcousticModel package but I’ve struggled to find any information on preparing Julius-friendly dict & dfa files. Can you point me in the right direction? Kind Regards, Oko http://www.voxforge.org/home/forums/message-boards/general-discussion/use-of-the-quickstart-nightly-package-with-a-htk_acousticmodel-nightly Wed, 27 Feb 2008 05:40:35 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-2008 What is Google Summer of Code? http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-2008 Mon, 25 Feb 2008 13:47:55 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/submit-your-speech-and-win-ipod-touch I just wonder do we know about this initiative: http://www.voice2type.com/submit_speech https://sourceforge.net/forum/message.php?msg_id=4752651 http://www.voxforge.org/home/forums/message-boards/general-discussion/submit-your-speech-and-win-ipod-touch Thu, 31 Jan 2008 06:56:45 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/forum-post-rating Hi everyone, In order encourage new users to submit questions/comments to VoxForge, we've decided to remove the "thumbs down" rating for posts in all the forums. You can still "thumbs-up" a post to give quick positive feedback. But if you don't like what was written, you can either post a reply to offer some constructive feedback or just ignore it. thanks, Ken http://www.voxforge.org/home/forums/message-boards/general-discussion/forum-post-rating Mon, 31 Dec 2007 20:32:07 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/language-hebrew Hello, First of all, great job with this project! I think that this project will be one the most popular projects in the open source world! I searched the forums a little bit, and found nothing regards Hebrew, so I guess no one worked on Hebrew yet ... Therefore, I willing to help open and maintain a section for Hebrew (with everything needed, speech recordings etc). However, I really don't know what I need to do in order to start such thing, so every help will be appreciated. Regards, Ofir http://www.voxforge.org/home/forums/message-boards/general-discussion/language-hebrew Sun, 23 Dec 2007 03:16:16 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/testing-nightly-build-of-accustic-models Hey everybody, i can't figure out how to setup a grammar that is based on the latest nightly builds of the acoustic models. For testing i have configured a the following voca: % NS_B <s> sil % NS_E </s> sil % COMMAND_START_V TIME t ay m sp DEPARTURE d ix p aa r ch er sp where i have configured two words that i selected from the dict file. here's the grammar: S : NS_B SENT NS_E SENT: COMMAND_START_V if i start julian i get some error msgs regarding missing phones: ###### check configurations ###### initialize input device ###### build up system Reading in HMM definition...(ascii)...limit check passed defined HMMs: 6757 logical names: 8564 in HMMList base phones: 44 used in logical done Making pseudo bi/mono-phone for IW-triphone...1062 added as logical...done reading [grammar/nabaztag.dfa] and [grammar/nabaztag.dict]... Reading in dictionary... line 3: triphone "ay-m+sp" not found line 3: triphone "m-sp+*" or biphone "m-sp" not found > 2 [TIME] t ay m sp line 4: triphone "ch-er+sp" not found line 4: triphone "er-sp+*" or biphone "er-sp" not found > 2 [DEPARTURE] d ix p aa r ch er sp ////// Missing phones: ay-m+sp ch-er+sp er-sp+* or biphone er-sp m-sp+* or biphone m-sp ////////////////////// error in reading grammar/nabaztag.dict: 2 words failed out of 4 words which seems to be correct as they are not listed in the tiedlist file. can anybody give me a hint why these phones are missing? the master prompts files still lists some sentences using both "time" and "depature" sebastian http://www.voxforge.org/home/forums/message-boards/general-discussion/testing-nightly-build-of-accustic-models Mon, 03 Dec 2007 15:57:07 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/gender--ts I would like to contribute to this project, however the first question is what gender am I? Well I am a transsexual and do not ascribe to either, and would still like to help by contributing to the voice sample project but am excluded How about being inclusive? http://www.voxforge.org/home/forums/message-boards/general-discussion/gender--ts Mon, 03 Dec 2007 07:48:30 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/make-sure-your-audio-editor-uses-libflac-version-1_2_1 From research.eeye.com: Multiple Vulnerabilities in .FLAC File Format and Various Media Applications Overview: eEye Digital http://www.voxforge.org/home/forums/message-boards/general-discussion/make-sure-your-audio-editor-uses-libflac-version-1_2_1 Mon, 19 Nov 2007 22:08:41 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/language-recognition Hi, congratulations for this valuable project! I'd like to start experimenting with voice processing in order to build an automated language recognition engine. I'm very new to this area but I can imagine a simple processing system that performs phoneme extraction from voice file, and then use an n-gram regognition system. Is the "accoustic model" of a language the right starting point for my approach? Vincent http://www.voxforge.org/home/forums/message-boards/general-discussion/language-recognition Mon, 19 Nov 2007 11:38:22 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/childes/talkbank Sorry if you know this link already. It seems missing on this site. I've just discovered our friends: http://talkbank.org/ - transcribed adults conversation under GPL http://childes.psy.cmu.edu/ - childrens under GPL http://www.voxforge.org/home/forums/message-boards/general-discussion/childes/talkbank Sat, 20 Oct 2007 08:44:22 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/mojomove---librivox-community-podcast-site-and-forum MojoMove is a new site that houses Podcasts that sometimes get incorporated into the LibriVox community podcast feed. They also have a Forum http://www.voxforge.org/home/forums/message-boards/general-discussion/mojomove---librivox-community-podcast-site-and-forum Mon, 08 Oct 2007 13:05:15 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/software-freedom-day---saturday-september-15th Software Freedom Day is a global, grassroots effort to educate the http://www.voxforge.org/home/forums/message-boards/general-discussion/software-freedom-day---saturday-september-15th Fri, 14 Sep 2007 08:00:36 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/avios-student-programming-speech/multimodal-application-programming-contest The Applied Voice Input/Output Society (AVIOS) has announced their second student application contest. Applications must involve speech input and/or output, but may be pure speech or multimodal. Cash and/or equipment prizes valued at over $1000 will be awarded to teams of student programmers who design and create applications judged by industry experts to be the most robust, useful, creative, innovative, and user friendly. The contest encourages students to develop applications using speech technologies such as automatic speech recognition and text to speech synthesis and to combine them with other modalities. This year students may use any of a variety of platforms, including Microsoft SAPI 5.3 in Windows Vista, CMU's RavenClaw/Olympus, Opera's X+V, Speech Application Language Tags (SALT), Voxeo Prophecy, as well as any of several on-line VoiceXML development environments (BeVocal Cafe, Loquendo Cafe, TellmeStudio, VoiceGenie Developer Workshop, and Voxpilot Voxbuilder). AVIOS president K.W. (Bill) Scholz explains: "Students will build creative and innovative applications that will lead the speech industry forward into new areas. The contest also provides a forum for students to show what they can do with the power of speech applications." Results from last year’s contest and more information about this year’s contest are at http://avios.com/contest.htm http://www.voxforge.org/home/forums/message-boards/general-discussion/avios-student-programming-speech/multimodal-application-programming-contest Wed, 12 Sep 2007 17:10:33 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/new-submissions-will-be-covered-under-gpl-v3 Just a note to let you all know that we've changed the license on the VoxForge site to GPLv3. Therefore, any new speech submissions to the VoxForge site will now be covered under GPLv.3. Since all speech submitted to VoxForge thus far included this notice: http://www.voxforge.org/home/forums/message-boards/general-discussion/new-submissions-will-be-covered-under-gpl-v3 Wed, 22 Aug 2007 20:56:04 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/speech-submission-feedback Here is a thread with respect to another user's (very valid) opinion as to the state of the VoxForge Submission System: http://www.voxforge.org/home/forums/message-boards/general-discussion/speech-submission-feedback Sun, 22 Jul 2007 21:41:16 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/text-to-speech can you tell me please how to make txt files into mp3 files of greek language? a list of tools and voices that can do this thanks http://www.voxforge.org/home/forums/message-boards/general-discussion/text-to-speech Fri, 06 Jul 2007 01:47:43 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/speech-synthesis-using-acoustic-model Hi, I am wondering if it is possible to use the HMMs of the trained Acoustic Model to synthesize speech. It should be possible to generate the most likely output sequence of MFCC frames given any input sequence of phonemes. Would this synthesized speech resemble the voice of the speaker who trained the AM (assuming that the AM was trained by a single speaker)? Maybe this is the standard synthesizer method in combined speech recognition & synthesis tools? Can someone point me to examples, possibly with technical description of the synthesizer? If this approach is not recommended, why (bad speech quality, waste of computing time,...)? Thanks John http://www.voxforge.org/home/forums/message-boards/general-discussion/speech-synthesis-using-acoustic-model Mon, 02 Jul 2007 05:41:14 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/cmu-language-modeling-toolkit-installing-in-cygwin Hi, I tried to install CMU language modeling toolkit in cygwin, but everytime I got error message. Does anyone know how to install it in cygwin? Best regards, Abdul. http://www.voxforge.org/home/forums/message-boards/general-discussion/cmu-language-modeling-toolkit-installing-in-cygwin Tue, 19 Jun 2007 04:56:54 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/new-logo Many thanks to Zachary Whitley for submitting a stylized SVG version of the old VoxForge logo, which I then converted to 3D using GIMP. Ken http://www.voxforge.org/home/forums/message-boards/general-discussion/new-logo Wed, 30 May 2007 09:35:27 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/unclear-cause-of-errors-when-using-trigram-lm-in-julius Hello, I have been trying to use HMM models that I've created with HTK and an ARPA LM in Julius (version 3.5.3, multipath compile option enabled, under Linux). Although the LM works without problems in HTK, Julius generates many warnings when parsing the LM and, finally, an error. All warnings are of this type: > Warning: context (z_y:_t_,z_j_@_n_) not exist in LR 2-gram (ignored) and this is the error: > Error: 2-gram has no upper 3-gram, but not 0.0 back-off weight What I don't understand is why HTK doesn't complain and Julius generates many warnings and an error. Does anyone know what these warnings mean? In other words: what is wrong with the LM, and why does Julius complain? I am loading the bigram and trigram LMs with the -nlr and -nrl options. The trigram LM file also contains the unigram and bigram LM. I found out that if I remove all unigrams and bigrams from the trigram file, Julius *does* start up in interactive mode. However, Julius gives an extra warning: > Reading in RL 3-gram... > Warning: 1-gram total num differ! may cause read error > Warning: 2-gram total num differ! may cause read error > reading 1-gram part... I am not sure if this warning is important or not. Can the results be trusted if I start Julius in this way? I have made one modification to the LM I used for HTK, in order to get it working in Julius. The LM used for HTK was in so-called "modified ARPA" (see HTK book) format, in which the back-off weights are optional. Julius doesn't (seem to) support this, so I filled in '0' everywhere a back-off weight was required but not filled in. Is this a good thing to do? Additionally, Julius doesn't load all trigrams: it stops after loading about 60% of all trigrams. Does anyone know why Julius would do this? Is it possible to find out if a triphone is responsible for stopping the loading? > 3-gram read 2500000 (57%) > <cut> > 3-gram read 2592555 end I would really appreciate your help! Best regards, Wout Output from Julius: include config: conf.jconf ###### check configurations ###### build up system Reading in HMM definition...(ascii)...limit check passed defined HMMs: 6849 logical names: 130051 in HMMList base phones: 51 used in logical done Making pseudo bi/mono-phone for IW-triphone...5150 added as logical...done Reading in dictionary... 4996 words...done Reading in LR 2-gram... reading 1-gram part... 1-gram read 4996 end reading 2-gram part... 2-gram read 0 (0%) 2-gram read 100000 (7%) 2-gram read 200000 (15%) <some lines removed> 2-gram read 1300000 (99%) 2-gram read 1306905 end done Reading in RL 3-gram... reading 1-gram part... 1-gram read 4996 end reading 2-gram part... Warning: (E_@_l_,2:_) not exist in LR 2-gram (ignored) Warning: (E_@_n_,2:_) not exist in LR 2-gram (ignored) Warning: (r_o:_,2:_) not exist in LR 2-gram (ignored) Warning: (t_s_e:_,2:_) not exist in LR 2-gram (ignored) <removed 100.000 lines> Warning: (z_a_x_,z_y:_t_) not exist in LR 2-gram (ignored) Warning: (z_i:_p_,z_y:_t_) not exist in LR 2-gram (ignored) Warning: (z_u:_,z_y:_t_) not exist in LR 2-gram (ignored) Warning: (z_y:_,z_y:_t_) not exist in LR 2-gram (ignored) 2-gram read 1306905 end reading 3-gram part... 3-gram read 0 (0%) Warning: context (2:_,Q_u:_6:_) not exist in LR 2-gram (ignored) Warning: context (2:_,Q_u:_6:_) not exist in LR 2-gram (ignored) Warning: context (2:_,Q_u:_6:_) not exist in LR 2-gram (ignored) Warning: context (2:_,d_i:_) not exist in LR 2-gram (ignored) Warning: context (2:_,r_) not exist in LR 2-gram (ignored) Warning: context (6:_,@_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) Warning: context (6:_,Q_I_u:_) not exist in LR 2-gram (ignored) <removed 1.500.000 lines> Warning: context (z_y:_t_,z_aI_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_aI_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_e:_6:_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_i:_t_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_j_@_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_j_@_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_j_@_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_j_@_n_) not exist in LR 2-gram (ignored) Warning: context (z_y:_t_,z_o:_n_) not exist in LR 2-gram (ignored) 3-gram read 2592555 end Error: 2-gram has no upper 3-gram, but not 0.0 back-off weight Terminated http://www.voxforge.org/home/forums/message-boards/general-discussion/unclear-cause-of-errors-when-using-trigram-lm-in-julius Wed, 23 May 2007 08:17:48 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge----ubuntu-collaboration Hello, I'm the accessibility coordinator for the Ubuntu project. I'd like to see some progress made on the Linux speech recognition front and I think the VoxForge initiative is a great way to start. Pooling resources is the only way to go! I see a few ways we may collaborate. For a start we could work together on GSoC projects. We have a fairly good track record with Google, having mentored about 20 projects each year from 2005. I did two myself in the accessibility field last year and will do three this year. Accessibility is such a narrow field that I feel several projects like Ubuntu, VoxForge, Orca and eSpeak should consider making a common project application to Google for 10-15 places. (btw, I realise that VoxForge is only partly about accessibility) Several of the projects listed in your GSoC forum section would be suitable for collaboration between our projects, and certainly the voice recording client. If we distribute that with Ubuntu (in universe at least) we might see decent participation numbers. It would be great if the same application also facilitated auditing of text-to-speech output. Ubuntu 7.04 just shipped with the eSpeak TTS engine with a handful of languages, but most of them could use some work. I would think parts of that could be recycled in the Dialog Manager GUI as well. I'm currently working on a specification for the speech recognition front-end and will post a link here once I've completed the first draft of it. Henrik http://www.voxforge.org/home/forums/message-boards/general-discussion/voxforge----ubuntu-collaboration Thu, 26 Apr 2007 18:33:58 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/dictation hello I am interested to speech dictation for greek is there a program to do it? thanks http://www.voxforge.org/home/forums/message-boards/general-discussion/dictation Mon, 23 Apr 2007 10:32:53 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/simon-dialog-manager-and-julian-speech-recognition My post to the Simon SourcForge Forum Hi bedahr, http://www.voxforge.org/home/forums/message-boards/general-discussion/simon-dialog-manager-and-julian-speech-recognition Thu, 19 Apr 2007 10:24:08 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/-german-speech-recognition-suite-gpl Post from Peter Grasch (see original post): http://www.voxforge.org/home/forums/message-boards/general-discussion/-german-speech-recognition-suite-gpl Mon, 16 Apr 2007 11:26:01 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-voice-local-search http://www.voxforge.org/home/forums/message-boards/general-discussion/google-voice-local-search Sat, 07 Apr 2007 11:37:48 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/asterisk-based-user-speech-submission-system Submission by trevarthan (see original post here) http://www.voxforge.org/home/forums/message-boards/general-discussion/asterisk-based-user-speech-submission-system Wed, 04 Apr 2007 10:36:59 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/corpus-thresholds Is there already some statistics about the actual size of Corpus being built and an estimate of the "distance" to a working threshold? http://www.voxforge.org/home/forums/message-boards/general-discussion/corpus-thresholds Tue, 03 Apr 2007 11:44:14 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/recognise-one-sentence-and-save-it__ My purpose of using speach recognition is to recognise one sentence, sending the result to another programfile and then exit. I would like to call julian using a systemcommand or some equal from a c-file (OS is linux) and then recognise one sentence, which result is sent to the c-file or saved to a textfile I can load from the c-file afterwards. I am a litte unsure about how to do this, and need help. My surgestion is to generate the grammar and record a voice-file with a matching sentence. Then execute julian from the c-file and make it listen to the voicefile to adjust input. Then it should record voice from a mic and try to recognise input. After this it should send the result or write it to a textfile and then exit.. How can I do this..? - Grant P.s. Sorry if it may have doubleposted this question, it did not seem to work the first time.. http://www.voxforge.org/home/forums/message-boards/general-discussion/recognise-one-sentence-and-save-it__ Wed, 28 Mar 2007 04:33:43 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-recognise-one-sentence-and-then-exit2 My purpose of using speach recognition is to recognise one sentence, sending the result to another programfile and then exit. I would like to call julian using a systemcommand or some equal from a c-file (OS is linux) and then recognise one sentence, which result is sent to the c-file or saved to a textfile I can load from the c-file afterwards. I am a litte unsure about how to do this, and need help. My surgestion is to generate the grammar and record a voice-file with a matching sentence. Then execute julian from the c-file and make it listen to the voicefile to adjust input. Then it should record voice from a mic and try to recognise input. After this it should send the result or write it to a textfile and then exit.. How can I do this..? - Grant http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-recognise-one-sentence-and-then-exit2 Wed, 28 Mar 2007 04:31:20 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-recognise-one-sentence-and-then-exit My purpose of using speach recognition is to recognise one sentence, sending the result to another programfile and then exit. I would like to call julian using a systemcommand or some equal from a c-file (OS is linux) and then recognise one sentence, which result is sent to the c-file or saved to a textfile I can load from the c-file afterwards. I am a litte unsure about how to do this, and need help. My surgestion is to generate the grammar and record a voice-file with a matching sentence. Then execute julian from the c-file and make it listen to the voicefile to adjust input. Then it should record voice from a mic and try to recognise input. After this it should send the result or write it to a textfile and then exit.. How can I do this..? - Grant http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-recognise-one-sentence-and-then-exit Wed, 28 Mar 2007 04:28:54 -0500 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-application Hi, The Google Summer of Code Mentor Application site opened Monday (March 5). I http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code-application Tue, 06 Mar 2007 00:03:02 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-can-i-make-new-words__ Hi.. I'm working on a project using Julian for speechrecognition and have followed the instruction regarding the grammar, but cannot find out how, if possible, I can create new words not listed in the dictionary-file/lexicon downloaded. I have created the grammarfiles but do not know how I can recognice these words. When I make the acoustic Model I will declare the pronaunsation of them, but compiling the grammar it just complaining about errors regarding these words, how can I help this..? Daniel http://www.voxforge.org/home/forums/message-boards/general-discussion/how-can-i-make-new-words__ Thu, 22 Feb 2007 04:51:00 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code http://www.voxforge.org/home/forums/message-boards/general-discussion/google-summer-of-code Fri, 16 Feb 2007 13:14:58 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/spice-speech-processing---interactive-creation-and-evaluation-toolkit-for-new-languages http://www.voxforge.org/home/forums/message-boards/general-discussion/spice-speech-processing---interactive-creation-and-evaluation-toolkit-for-new-languages Tue, 13 Feb 2007 09:42:43 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-get-many-more-contributions If you'd like to encourage many more contributions, how about developing a Macromedia Flash voice recorder and embedding it in your website? This could make it really quick and easy for people to contribute, and persuade many "casual visitors" to record a few of the scripts. Cheers, Jon (www.orangejon.com) http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-get-many-more-contributions Thu, 18 Jan 2007 18:48:25 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/speech-recognition-engine-comparison Hello All, I am just curious to know if it possible to build a good recognition system comparable to commercial ones by Philips, Nuance etc. by using Open source technologies like Juilus, Sphinx etc. I am working hard to get a ASR (Sphinx ) working with Hub4/WSJ but it seems to be a distant fruit. Any wikis or tutorials that help much better? Any one ready to work on a ground-breaking idea using a mash up of Web2.0 Technologies and ASR?? thanks Satish mummsat@iit.edu http://www.voxforge.org/home/forums/message-boards/general-discussion/speech-recognition-engine-comparison Wed, 03 Jan 2007 18:50:30 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/can-we-use-th-hub4-or-wsj-thats-supported-by-sphinx How do i use Hub4 or AN4 or WSJ models/dictionaries that are suported by sphinx? Can we use them with Julius? thanks Satish http://www.voxforge.org/home/forums/message-boards/general-discussion/can-we-use-th-hub4-or-wsj-thats-supported-by-sphinx Mon, 01 Jan 2007 17:24:28 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-create-reverse-3-gram-for-julius3 I am able to create a back-off bigram using HTK with Julius uses for its first pass. But I really have no idea how to create the REVERSE 3 gram needed for the second pass. Could anyone shed some light? Thanks. http://www.voxforge.org/home/forums/message-boards/general-discussion/how-to-create-reverse-3-gram-for-julius3 Wed, 20 Dec 2006 23:11:54 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/gpl-applicable-to-derived-hardware Hi, I wonder, how the GPL would be applied to the following situation: A huge number of users like you and me collect a speech corpus that allows to build acoustic models of similar quality compared to commercial systems. Now a hardware developer comes up with some speech recognition tool that is designed to work exactly with VoxForge's model. He doesn't distribute the model with his device, but provides a link to voxforge.org where customers may download the model. - Is this situation covered by GPL? If not, I'm reluctant to spend time on something that others may turn into profit without returning anything to the community. If yes: - Does the GPL require the device developer to distribute his device together with the model sources, OR - Does the GPL reuqire him to distribute his device together with the model sources PLUS any sources and hardware description of his own device? The latter situation is most desirable, because everybody who benefits from the acoustic models has to provide some improvement in return. But I doubt GPL includes anything except for directly derived code (or models). Your opinion? http://www.voxforge.org/home/forums/message-boards/general-discussion/gpl-applicable-to-derived-hardware Sun, 17 Dec 2006 16:10:58 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/beginners-questions Hi, congratulations for this valuable project! I'd like to start experimenting with HTK and have a few questions. 1) Where do I find english documentation for Julius? On the Japanese sourceforge site first of all I see a lot of '?????'. Is it an alternative to HTK, or do I need to install it? 2) Will there be any compiled Acoustic Models (trained HMMs)? This is what I thought to find below downloads->acoustic models, instead there are word-phoneme dictionaries. Probably I am misunderstanding something? 3) If I submit speech, do I just have to provide a text transcription, or also a phonetic transcription? What list of phonemes do you use? The format of the text transcriptions seems to differ between the corpora, which rules should I follow? 4) Is there a standard, how the MFCC coefficients are calculated? There are a lot of options concerning frequency bands, triangular/hamming filter functions in mel/linear space etc. Arno http://www.voxforge.org/home/forums/message-boards/general-discussion/beginners-questions Fri, 15 Dec 2006 13:51:05 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/svn-checkout Where can I svn co the repo I see at http://www.dev.voxforge.org/browser/VoxForge/Trunk http://www.voxforge.org/home/forums/message-boards/general-discussion/svn-checkout Mon, 06 Nov 2006 00:43:25 -0600 http://www.voxforge.org/home/forums/message-boards/general-discussion/how-do-someone-create-language-model-julius/julian Hi All, http://www.voxforge.org/home/forums/message-boards/general-discussion/how-do-someone-create-language-model-julius/julian Sun, 22 Oct 2006 13:37:03 -0500