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
3
votes
2 answers

Build sphinx4 using eclipse

I have been trying to build sphinx4 in eclipse for a long time with no success! I import all the sphinx sources and add all the necessary external Jar files, but eclipse doesn't recognize most of the packages therefore I can't build it. Many…
user577295
  • 31
  • 1
  • 3
3
votes
1 answer

Error at ARPA model training with SRILM

I have followed this tutorial. After I run this code: ngram-count -kndiscount -interpolate -text train-text.txt -lm your.lm It gives me this error: "One of modified KneserNey discounts is negative error in discount estimator for order 2." How…
ziLk
  • 3,120
  • 21
  • 45
3
votes
0 answers

How to load Transform from a file in to Recognizer in Sphinx4-5prealpha

Usaing sample code TranscriberDemo.java I successfully save Transform in to a file as follows: // Transform represents the speech profile Transform transform = stats.createTransform(); transform.store("MyVoiceTransform", 0); Now I want to use it in…
g10dras
  • 399
  • 2
  • 11
3
votes
2 answers

Error using sphinx4 jars without Maven

I have a problem with the API Sphinx4 and I can't figure out why it doesn't work. I try to write a little class for capture the voice of an user and write his speaking on a file. 1) I have create a new java project on Eclispe. 2) I have create the…
tirz
  • 2,041
  • 1
  • 22
  • 37
3
votes
1 answer

How do I tell the difference between an 8Khz acoustic model and a 16Khz model?

I'm able to get a reasonable level of accuracy with 8khz audio files. Now I want to try a higher sample rate, if I can. Looking at the acoustic models available on this page, they list: en-us-8khz.tar.gz en-us-semi-full.tar.gz …
user773737
3
votes
1 answer

What do I do if sphinx is completely inaccurate?

Edit for clarity: The output I get from sphinx is not even close to the actual words in my sound file. What do I need to do to make it more accurate? Here's the file I'm trying to get a transcript from. It should be at a sample rate of 8K. Acoustic…
user773737
3
votes
1 answer

CMUSphinx live speech recognition too slow?

CMU Sphinix is toooo Slow for recognizing live speech.I don't know if you have any idea for boost it? This is my…
Xmac
  • 53
  • 6
3
votes
1 answer

Keyword or keyphrase spotting with Sphinx4

I am currently trying to make my java code (using eclipse) perform some function if a certain thing is said. I am using the Sphinx4 libraries and this is what I currently have: What I would like it to do is at the line where it says: IF (TRUE)…
mray190
  • 496
  • 3
  • 13
3
votes
1 answer

Java Speech API tutorials?

I'm looking to play around with the JSAPI, but I can't find any tutorials on how to do it. I know Sphinx 4 uses JSAPI and there are a few (pretty bad) tutorials for that. From preliminary research and playing around with Sphinx, it's a lot more than…
rphello101
  • 1,671
  • 5
  • 31
  • 59
3
votes
2 answers

Extending Sphinx to support a local language

Initially i was thinking of writing a speech recognition engine from scratch(with support of 50-100 words)to support my native language. However after some research it has become clear that it is not possible to make a speech recognition engine even…
Win Coder
  • 6,628
  • 11
  • 54
  • 81
3
votes
1 answer

Tuning CMU Sphinx

I am trying to build a speech recognition app based on CMU Sphinx. I have created my own language model using the lmtool. But inorder to improve the accuracy of recognition, I want to tune Sphinx. But are there any guidelines for choosing the…
Shishya
  • 1,069
  • 1
  • 14
  • 22
3
votes
2 answers

How to implement pocketsphinx with windows 7 and eclipse ?Or how to use java sphinx speech recognition to android?

I want To use sphinx 4 speech recognition in android.Recently I used sphinx4 speech recognition for java and implemented successfully. But my problem is pockesphinx is speech recognition for android.I could not understand the tutorial.I tried so…
BhavikKama
  • 8,566
  • 12
  • 94
  • 164
2
votes
2 answers

Speech recognition response is poor in sphinx4

Currently we are investigating into using sphinx4 for speech recognition. We are trying to achieve a good response for a dictation type application. The input is a wav file and we wish to transcribe it. I have looked into the LatticeDemo and…
Raveesh Sharma
  • 1,486
  • 5
  • 21
  • 38
2
votes
2 answers

How large must a corpus be to create a language model for Sphinx?

I would like to know how many documents or sentences or words I need to process in order to get a good language model of a domain and use it in voice recognition tools such as CMU Sphinx.
pjvv1
  • 21
  • 2
2
votes
0 answers

Unable to recognise the Live stream voice using sphinx4

I have created code based on this tutorial Performing Speech to text from a .wav file using StreamSpeechRecognizer works fine but I am facing an issue with LiveSpeechRecognizer, In debug the code goes till recognizer.startRecognition(true);…
1
2
3
16 17