Questions tagged [sphinx4]

Sphinx-4 is a part of CMUSphinx Speech Recognition Toolkit. It's a flexible large and small vocabulary speech decoder written in Java and licensed under BSD license.

This tag is about Sphinx-4, a speech recognition decoder. Speech recognition is fastly growing domain and it's quite complex by its nature. The development of the speech recognition application requires understanding of the speech recognition specifics - a probabilistic nature of the results, the need for throughout testing, the specifics of the voice user interface design, the accuracy/speed balance.

The main concept you need to be aware about are acoustic model used to capture the sounds of the language, the language model used to capture the vocabulary and the dictionary which maps from words to sounds. The use of Sphinx-4 in your application is often straightforward but you need to be more careful than usually to get everything in place.

To learn more about CMUSphinx and Sphinx-4 visit CMUSphinx page

https://cmusphinx.github.io/wiki/

Read the tutorial

https://cmusphinx.github.io/wiki/tutorial/

255 questions
1
vote
1 answer

Sphinx4: recognize() won't run the second time

i am using sphinx4 to pause a thread until a specific keyword is spoken. This works great the first time, but the second time i need to pause the thread, the recognizer.recognize() seems to not run and the application just starts spamming "Start…
user1419305
  • 468
  • 8
  • 16
1
vote
1 answer

Configuration manager of Sphinx4 returns null from lookup

i encountered a problem while following this tutorial to create a speech recognition using Sphinx4 with the Java-Speech-API. I did the exactly same as the author did in his blog, but when i execute my program i receive a NullPointerException…
LorisBachert
  • 283
  • 1
  • 2
  • 12
1
vote
1 answer

how to set minimum acceptable recognition accuracy?

I need to recognize key words in long speech audio file (wav). There can be different noise, clicks, clangs, etc. This sounds are often recognized as keywords. How can I set some threshold to distinguish words correctly?
1
vote
0 answers

Recognizer not working efficiently after adding more words in grammar of sphinx 4

i have increased the size of grammar and hence added many more words in the dictionary with different accents by using different HMMs. But the recognizer is not working efficiently and taking absurd words or may not respond. It is happening mostly…
utsav
  • 71
  • 1
  • 7
1
vote
1 answer

microphone not starting after one iteration in sphinx

I am using CMU sphinx library, but after successfully running one iteration, this error appears "Can't open microphone line with format PCM_SIGNED 16000.0 Hz, 16 bit, mono, 2 bytes/frame, big-endian not supported." Cannot start microphone. When run…
utsav
  • 71
  • 1
  • 7
1
vote
2 answers

Decode speech into Phonemes in Sphinx4

Can I use CMUSphinx4 to decode a particular speech into phonemes and get those phonemes into further implementations?
Sameera
  • 304
  • 1
  • 19
1
vote
1 answer

Sphinx 4: Same inputs produce different results each time

I want to make Sphinx4 deterministic. Each time I give the same input files in my unit tests, I want it to recognize the same words (with same timestamps). I've tried looking through the source code and fixing all calls to new Random(), but that…
Aleksandr Dubinsky
  • 22,436
  • 15
  • 82
  • 99
1
vote
1 answer

Error running helloworld sphinx

Hi i am making a software in java in which i want to develop a speech software... I am running a "Hello" sphinx code in java: import edu.cmu.sphinx.frontend.util.Microphone; import edu.cmu.sphinx.recognizer.Recognizer; import…
karensantana
  • 1,599
  • 4
  • 21
  • 34
1
vote
1 answer

How to add keywords that the user gets to pick in Sphinx 4 (Voice Recognition)?

I was wondering how to add custom keywords to voice recognition using sphinx 4 in Java using the GRAM file? Like: public = (set song); How would I make this so after 'set song' the user could say anything he wants and it would still get parsed? …
1
vote
1 answer

Sphinx 4 Integrate language model

I am new to Sphinx 4. I have been reading about Sphinx 4 and have to implement it as a part of the project. I am building upon HelloNGram model. To improve accuracy I wish to include a language model I downloaded from…
Shikhar Shrivastav
  • 725
  • 2
  • 11
  • 21
1
vote
1 answer

Sphinx4 speech recognition trasncribe demo not working accurately for short wav file

i had just implemetned the transcriber demo for the transcribe the audio file .. My audio file is .wav file which consist only names like "BHAVIK" "ANKIT" "SAGAR" My grammer File Consist of this grammer as follows: - public =…
BhavikKama
  • 8,566
  • 12
  • 94
  • 164
1
vote
1 answer

Sphinx4 to recognize the speech from the wav file as input rather then microphone input

I had implemented the sphinx4 speech recognition in java successfully..now i just wanna use that sphinx 4 speech recognition with the ready made .wave file. right now i just speak and its recognize the words as per ma grammar.. but how can i just…
BhavikKama
  • 8,566
  • 12
  • 94
  • 164
1
vote
0 answers

Display phonemes instead of text [pocketsphinx]

I have successfully compiled the HelloWorld tutorial of PocketSphinx on my Android device. [link : http://cmusphinx.sourceforge.net/2011/05/building-pocketsphinx-on-android/] Now, I need some help to tweak the code of the HelloWorld tutorial so that…
user782400
  • 1,617
  • 7
  • 30
  • 51
1
vote
1 answer

Bias while converting CMU Sphinx's confidence score to probability

I am trying to convert output of CMU Sphinx's recognizer (i.e. list < hypothesis (i.e. phrase), score (in log) > obtained by tweaking test_ps_nbest.c) to following form: list < hypothesis (i.e. phrase), "probability" (between 0 and 1) > A trivial…
1
vote
1 answer

Can record sound using Microphone once, but cannot start microphone second time

I am using the CMU sphinx library to record sound. When i begin my java application, I allocate the Recognizer and the Configuration Manager only once as follows: cm = new…
Programmer
  • 6,565
  • 25
  • 78
  • 125