I am sorry that you are having problems with the software. There are any number of reasons why things might not work, but you did not provide detail on what you are doing.so it's not possible to say anything. You reached out through Sourceforge, which has not been the active repository for a while. http://cmusphinx.github.io/ hosts the community of current sphinx users; it has a forum where you can ask for help. You should use the github repository and its current software. You can get help through...
The software has evolved and is currently hosted at http://cmusphinx.github.io/ The sorxceforge version is not currently maintained. The github site also has a user forum where you can ask about any issues. On Tue, Jan 9, 2024 at 12:27 AM Korey F Schulz foxtrot1@users.sourceforge.net wrote: Hi. I am updating the model for an application that I am building but I can't make progress in the tutorial because none of the utilities that are referenced exist on my machine. The missing utilities include...
Please look at http://cmusphinx.github.io/ On Sat, Jan 6, 2024 at 9:00 AM HACHICHA YOUSSEF yousseftakihach@users.sourceforge.net wrote: I dont know what it is but no matter what file i give it either a video or audio the transcription is not accuarate at all, even tho that the files are clear english it's really bad. do i have to train it? if yes can someone tell me how to do it in python? that's my code bellow ↓↓↓ also its 2024 arent you planning on changing the conversation over to slack or some...
Ben, This would depend on the acoustic model. If you have (or have trained) such a model, and have a corresponding dictionary. It should be possible to do. cmusphinx does not prescribe a phone set, so IPA would work fine. There is work on Czech ASR you could look up and see what rhey did. Best, Alex On Tue, Dec 5, 2023 at 12:36 PM "Benjamín Novický" brenko@users.sourceforge.net wrote: Hello, I am working on my bachelor thesis which is training a cusphinx model for speech recognition in Czech language....
Current speech speech systems can use speech-to-letter models for decoding the audio signal, together with a language model to detect legal sequences. Look into into speech-to-vec 2 models; thesecan generate a feature space that can then be trained to classify tokens for specific languages. With either approach you will need to have at least some annotated speech to map the audio to symbols. One rule of thumb is ~50 instances for each symbol, but this can vary accordining to the end task. Trying...
CMUSphinx has dynamic language models. It also has class language models. Your different contexts can use dynamic lms; you have to precompile them.. The class language models can have tokens that expand to a word list (think [cityname] -> Tokyo, London,...). But the words have to already be in your global dict. On Wed, Oct 26, 2022, 02:05 Ken Schafer byteplayer@users.sourceforge.net wrote: Okay, I've done some searching on these forums and have found some similar questions, but nothing exactly like...
Folks, Thank you all for your patience! lmtool is back up and appears stable. Please try it at http://www.speech.cs.cmu.edu/tools/lmtool-new.html. lextoo should be up shortly. If you encounter any problems please post or email me directly. Thank you, Alex Rudnicky On Fri, Mar 25, 2022 at 8:52 AM Daniel Wolf lupomuc@users.sourceforge.net wrote: The LMTool site (http://www.speech.cs.cmu.edu/tools/lmtool-new.html) appears to be down. It would be great if it could be got up again! LM tool https://sourceforge.net/p/cmusphinx/discussion/help/thread/758ed876/?limit=25#6476...
Folks, The lmtool is close to being up for general use; I have it working internally and it's being tested. Thanks, Alex On Wed, Apr 27, 2022 at 5:13 AM Mina Huh mina1004h@users.sourceforge.net wrote: @air https://sourceforge.net/u/air/ When I am trying to call on the CMU Sphinx lmtool to get pronunciations for words using python script, I get this error message. You dictionary could not be constructed because could not upload word file500 Can't write to /var/www/cgi-bin/tools/logios/../../../www-htdocs/tools/product/1651049109_20533/8508.word:...
The code for quick_lm had been posted to the website. It's actually in the cgi folder. I will post the code to github as well. Alex On Tue, Apr 19, 2022 at 1:32 PM Daniel Wolf lupomuc@users.sourceforge.net wrote: It's been several years since I last used the tools, so I can't give you much feedback. I used them for two things: I seem to remember that the LM generation scripts were somehow hosted on the site, allowing me to download and study them. It's surprisingly hard to find good (and simple!)...
Folks, lmtool is on the road to recovery. The site and the computer were hacked. I'm in the process of reconstituting the website and tool. Unfortunately the damage to lmtool appears to have been made around 15 sept 2021 while the computer itself was disabled in early February. But the code has been recovered. What's missing is the web page interfaces. We're still looking for them in the backups. It's really close at this point (he claims). I apologize for slow progress. My time had been taken by...
cmudict is on github; the lm compiler is not (I agree it should be). The rest of the stuff is cgi code and also a binary for pronunciations. The latter components are old and need to be replaced.... On Mon, Mar 28, 2022 at 1:48 AM Daniel Wolf lupomuc@users.sourceforge.net wrote: Oh, that sucks. Good luck restoring the contents! Is there a public repo with the source code? I did some searching, but all links pointed to the website only. LM tool https://sourceforge.net/p/cmusphinx/discussion/help/thread/758ed876/?limit=25#94cb...
The machine was hacked (and stuff wpied). The system will be restored using other copies, but it's taking time. Apologies for the mess. Alex On Fri, Mar 25, 2022 at 8:52 AM Daniel Wolf lupomuc@users.sourceforge.net wrote: The LMTool site (http://www.speech.cs.cmu.edu/tools/lmtool-new.html) appears to be down. It would be great if it could be got up again! LM tool https://sourceforge.net/p/cmusphinx/discussion/help/thread/758ed876/?limit=25#6476 Sent from sourceforge.net because you indicated interest...
Yes it is down. Apologies! I will look at it though possibly only tomorrow morning. Thank you for the heads up, Alex On Thu, Jan 21, 2021 at 4:47 AM Karim katiki@users.sourceforge.net wrote: Hi, Since this morning, it looks like the lm-tool webpage is no longer accessible: http://www.speech.cs.cmu.edu/tools/lmtool-new.html Is it normal ? K. lmtool webservice no longer accessible ? https://sourceforge.net/p/cmusphinx/discussion/help/thread/7e65904d44/?limit=25#3e80 Sent from sourceforge.net because...
The page that returns with the results from the compilation has tags that should make it easy to pick out the tar file. Apart from just regexing it, something like beautifulsoup should make it easy. I'm note sure I make that obvious in the description... quick_lm will give you a language model but not the dictionary. I know. It's an issue. But the (rather old) code is licensed. And I haven't had the perceived time to stick in an open source pronunciation generator. Alex On Wed, Jul 3, 2019 at 4:06...
Judging from your log file, the cmn vector is being computed from utterance to utterance. Eventually it settles into a 3-utt loop, which is why you're seeing what you're seeing. This doesn't make sense since you have '-cmn current' in your config; it's acting more like 'prior'. Look at the code and check that the cmn logic is correct... The 'first utterance is garbage' problem happens in prior mode, when the default cmn values are just wrong for the utt at hand. On Wed, Jan 31, 2018 at 10:08 AM,...
fix
Clarification of license reference and some add...
major roll-up of updates and corrections from b...
edits for reported error.
minor fix
new batch of additions and corrections
updates...
more fixes...
updates
new words added; some editing
recent updates (from cgi tool)
additional scripts for maintenance
updates
more changes
ortho fix to dict
svn:// -> http://
update
words additions taken from cmudict tool; other ...
Dictionary update; new words, corrections. Some...
You need to use the processed copy of the dictionary. The default version, corresponding...
Maintenance changes to scripts.
Compilation of most recent dict, with many upda...
Minor fixes and clean ups
rollup of dict mods, including a branch from a ...
more changes. this is a running update.
new words, variant reduction, corrections
some more fixes...
fixes...
fixed typos, collation
fix version
fix
The server hiccuped. It will happen; if you email to the address on the page, it...
incorporated cmudict-new words, plus others; ge...
added summing of vowel counts
new working copy...
this is obsolete; will copy branch trunk/ here
some touch-ups
A new version of the cmudict, rolling up variou...
additional error checking and warnings.
corrections and additions based on user feedback.