Temple University Speech Recognition using Sphinx 4 (Ti Digits test) Jaykrishna shukla,Amir Harati,Mubin Amehed,& cara Santin Department of Electrical.

Slides:



Advertisements
Similar presentations
Session # 2 SWE 211 – Introduction to Software Engineering Lect. Amanullah Quadri 2. Fact Finding & Techniques.
Advertisements

DETAILED DESIGN, IMPLEMENTATIONA AND TESTING Instructor: Dr. Hany H. Ammar Dept. of Computer Science and Electrical Engineering, WVU.
Regression Methodology Einat Ravid. Regression Testing - Definition  The selective retesting of a hardware system that has been modified to ensure that.
Copyright A. Andrews, 2003 Regression Testing Department of Computer Science.
Software Quality Assurance Plan
Existing Documentation
Architecture and Real Time Systems Lab University of Massachusetts, Amherst An Application Driven Reliability Measures and Evaluation Tool for Fault Tolerant.
Heuristic Evaluation.
1 USING CLASS WEIGHTING IN INTER-CLASS MLLR Sam-Joo Doh and Richard M. Stern Department of Electrical and Computer Engineering and School of Computer Science.
ECE122 L17: Method Development and Testing April 5, 2007 ECE 122 Engineering Problem Solving with Java Lecture 17 Method Development and Testing.
Testing - an Overview September 10, What is it, Why do it? Testing is a set of activities aimed at validating that an attribute or capability.
Problem with Software Requirements are complex The client does not know the functional requirements in advance Requirements may be changing Technology.
Chapter 11: Testing The dynamic verification of the behavior of a program on a finite set of test cases, suitable selected from the usually infinite execution.
Software Testing Test Design and Implementation. Agenda Test Design Test Implementation Test Design Sources Automated Testing 2.
Some Voice Enable Component Group member: CHUAH SIONG YANG LIM CHUN HEAN Advisor: Professor MICHEAL Project Purpose: For the developers,
TH-OCR NK. content introduction go to next page background assumptions overall structure chart IPO for overall structure dataflow diagram of overall structure.
Testing. Definition From the dictionary- the means by which the presence, quality, or genuineness of anything is determined; a means of trial. For software.
1. Topics to be discussed Introduction Objectives Testing Life Cycle Verification Vs Validation Testing Methodology Testing Levels 2.
Software System Engineering: A tutorial
C HU H AI C OLLEGE O F H IGHER E DUCATION D EPARTMENT O F C OMPUTER S CIENCE Preparation of Final Year Project Report Bachelor of Science in Computer Science.
Temple University Goals : 1.Down sample 20 khz TIDigits data to 16 khz. 2. Use Down sample data run regression test and Compare results posted in Sphinx-4.
Verification and Validation Overview References: Shach, Object Oriented and Classical Software Engineering Pressman, Software Engineering: a Practitioner’s.
Introduction to Java August 14, 2008 Mrs. C. Furman.
Programming Lifecycle
Engineer Training RTRun – Real Time Software TJ8300 / TJ8500 RTRun – Real Time Software.
TESTING FOR THE RELIABILITY OF A SOFTWARE SARAT CHANDRA YADAVALLI CSC 532 TERM PAPER.
Software Engineering 2 Software Testing Claire Lohr pp 413 Presented By: Feras Batarseh.
Data Submittals to AQS Nate Herbst Southern Ute Indian Tribe.
Well-behaved objects Main concepts to be covered Testing Debugging Test automation Writing for maintainability Objects First with Java - A Practical.
Temple University Training Acoustic model using Sphinx Train Jaykrishna shukla,Mubin Amehed& cara Santin Department of Electrical and Computer Engineering.
Temple University Training Acoustic Models Using SphinxTrain Jaykrishna Shukla, Mubin Amehed, and Cara Santin Department of Electrical and Computer Engineering.
1 Introduction to Software Testing. Reading Assignment P. Ammann and J. Offutt “Introduction to Software Testing” ◦ Chapter 1 2.
FY 2011 Public Libraries Survey WebPLUS Software Training.
Chapter Title CHAPTER 3 RESEARCH METHODOLOGY Research Methodology This part of the study includes Project Design, Project Development and Operation,
Systems Development Life Cycle
Project Management. Introduction  Project management process goes alongside the system development process Process management process made up of three.
Basic structure of sphinx 4
ECE 8443 – Pattern Recognition ECE 8527 – Introduction to Machine Learning and Pattern Recognition Objectives: Elements of a Discrete Model Evaluation.
BIF713 Operating System Concepts MS Windows. Agenda 1. What is an Operating System (definition)? 2. Types of Operating Systems 3. Basic Operations: –
Objects First With Java A Practical Introduction Using BlueJ Well-behaved objects 2.1.
A New Approach to Utterance Verification Based on Neighborhood Information in Model Space Author :Hui Jiang, Chin-Hui Lee Reporter : 陳燦輝.
C HU H AI C OLLEGE O F H IGHER E DUCATION D EPARTMENT O F C OMPUTER S CIENCE Preparation of Final Year Project Report Bachelor of Science in Computer Science.
ECE 8443 – Pattern Recognition ECE 8527 – Introduction to Machine Learning and Pattern Recognition Objectives: Statistical Significance Hypothesis Testing.
Software testing techniques Software testing techniques REGRESSION TESTING Presentation on the seminar Kaunas University of Technology.
Temple University Summer Research Progress: Week 2 – Extraction of Data George Fava Department of Electrical and Computer Engineering Temple University.
REGRESSION TESTING Software Quality Engineering NC Zunaira Tariq Bese 19B Software Quality Engineering NC Zunaira Tariq Bese 19B.
#SummitNow Yes, I'm able to index audio files within Alfresco 2013 Fernando González @fegorama.
REGRESSION TESTING Audrius Čėsna IFM-0/2. Regression testing is any type of software testing that seeks to uncover new errors, or regressions, in existing.
Solvency II Tripartite template V2 and V3 Presentation of the conversion tools proposed by FundsXML France.
Regression Testing with its types
System Design, Implementation and Review
Operating System Concepts
ADE EDIS READ & Optimizer TRAINING Colorado Department of Education
Verification and Validation Overview
System Development Life Cycle (SDLC)
Materials Engineering Product Data Management (ePDM)
Lab 2: Isolated Word Recognition
JCL Standards #5 Company Name
Introduction to Software Testing
Inventory Planning (I/P) Process Flow
Production Planning (P/P) Process Flow
How to Use References Chapter 4.
Lab 3: Isolated Word Recognition
Sphinx Recognizer Progress Q2 2004
Department of Computer Science Regression Testing.
Regression Testing.
Computer Science The 6 Programming Steps.
Chapter 2. Problem Solving and Software Engineering
M. Kezunovic (P.I.) S. S. Luo D. Ristanovic Texas A&M University
Presentation transcript:

Temple University Speech Recognition using Sphinx 4 (Ti Digits test) Jaykrishna shukla,Amir Harati,Mubin Amehed,& cara Santin Department of Electrical and Computer Engineering Temple University URL:

Temple University: Slide 1 Introduction & Goals There were two main goals for the week 1.To run the short Ti digit test data through the cmu sphinx 4’s Regression test and get the results. 2.Compare the % error results for the Sphinx-4 with ISIP recognizer. 3. To train a new acoustic model using SphinxTrain 1.0

Temple University: Slide 2 Concept of Regression test Regression testing is any type of software testing that seeks to uncover software errors by partially retesting a modified program. The intent of regression testing is to provide a general assurance that no additional errors were introduced in the process of fixing other problems. Regression testing is commonly used to efficiently test the system by systematically selecting the appropriate minimum suite of tests needed to adequately cover the affected change. Common methods of regression testing include rerunning previously run tests and checking whether previously fixed faults have re-emerged.software testing

Temple University: Slide 3 Steps for initiating the regression test in sphinx 4 The first step was to create our own batch file from the provided Ti digits short test data. This was done manually by referencing each raw file one by one (it was time taking) the second step was to initiate the test from the command line. In this case we choose to do the flat unigram test A user can run the test on 10 different parameters. To get the menu for the parameter list ant -projecthelp The user than can initiate the test by typing: ant followed by the type of test of their choice

Temple University: Slide 4 Sample out put of a regression test Sample output for the word list test

Temple University: Slide 5 Sample out put of a regression test (cont) The summary of the results from running 10 sample data from TIDigit_short file In Sphinx [java] # Summary statistics [java] Accuracy: 4.348% Errors: 22 (Sub: 4 Ins: 0 Del: 18) [java] Words: 23 Matches: 1 WER: % [java] Sentences: 11 Matches: 0 SentenceAcc: 0.000% [java] Total Time Audio: 6.55s Proc: 2.15s Speed: 0.33 X real time [java] Mem Total: Mb Free: Mb [java] Used: This: 9.23 Mb Avg: Mb Max: Mb Error rate is way high..is this due to Small data base

Temple University: Slide 6 The Flow chart for the Training Procedure Training acoustic model using SphinxTrain 1.0

Temple University: Slide 7 Detailed Training procedure The sphinxTrain 1.0 has a total of 12 steps The fire The to step we covered this week were 1. creating the model definition file 2. creating the HMM topology file. The major problem was that the documentation for SphinxTrain. The web documentation was out dated and did not match with their software.