Yes, Because I need help from anywhere I can get :-(
what is Kaldi? and how does it improve the accuracy?
what is Kaldi? and how does it improve the accuracy?
All files are here for training sentences.
Hello everyone, I am a researcher, and I am making speech recognition system using CMU Sphinx. I have succesfully trained words of Sindhi Language with very low accuracy. Now I wanted to train complete sentences. I want help, that how to train complete sentences, how to make dictionary, transcript, fileids and other files for complete sentences training. Help will be highly appreciated.
I am making a speech recognition system using CMU Sphinx, I have achieved words recognition with very low accuracy. Now I want to train complete sentences. For that I am getting many errors while training. Can someone help me with that? Please its a request, if anyone can.
This is the only 1 error I am facing in phase 3. Phase 3: Check general format for the fileids file; utterance length (must be positive); files exist WARNING: Error in 'C:/sphinx/other/etc/other_train.fileids', the feature file 'C:/sphinx/other/feat/Female1/abid.mfc' does not exist, or is empty WARNING: Error in 'C:/sphinx/other/etc/other_train.fileids', the feature file 'C:/sphinx/other/feat/Male1/abid.mfc' does not exist, or is empty WARNING: Error in 'C:/sphinx/other/etc/other_train.fileids',...
I need it too :-(
Help will be highly appreciated
Hello Sir, I was trying to connect my pocketsphinx command with Python GUI so that on a Button Click of GUI the speech recognition may start working. I could not find any tutorial, Can you help me how to do it? pocketsphinx_continuous -inmic "yes" -hmm "C:\sphinx\other\output\other.ci_cont" -dict "C:\sphinx\other\output\other.dic" -lm "C:\sphinx\other\output\other.lm.DMP" How to run this command and gets its output in python gui, rather than running the above command in command prompt
Thank you Sir. My error's are now resolved, now I have 89% accuracy. :-)
Kindly see this log file, It gives these 3 error? Why are they coming. ERROR: "dict.c", line 185: Line 25: No pronunciation for word 'SIL'; ignored ERROR: "batch.c", line 225: Empty mfcc file ERROR: "batch.c", line 490: Failed to read MFCC from the file 'C:/sphinx/other/feat/Male3/abid.mfc'
Kindly see this log file, I don't understand what is the error.
I tried 5 words of sindhi language,(Kindly see attachment for it) I recorded individual wav's for each word, total 5 It had accuracy 60% Now when I extended words upto 24, and for each word I have 5 wav's of different people, it is not generating the files as it did before in 5 words attempt. Error's are in other.html file, it is also attacted. It says it cannot find .mfc file
The problem is while generating files automatically, it gives me error's. Can you help me resolve them? Or refer me a complete tutorial to generate .DMP and .dic files I have generated previously files but with 1 wav 1 word, now I want to generate 1 word with multiple wav files of same word.
I am trying to create a language model for speech recognition of Sindh Language. I achieved single word single speech recognition with single audio sample. Now when I increased the number of words, the accuracy level degraded. Now I have multiple recordings of same word, I want to train each word with multiple audio samples for better accuracy, but I am facing some errors while generating files. If someone could help me please ? Note : I have uploaded all the files. Getting error while executing...